Future transaction information display program

Abstract

To provide valuable advice for a future transaction without requiring a special skill or experience.SOLUTION: A program makes a computer execute: an information acquisition step for acquiring demand information regarding demand for noble metal during trading; and a display step for displaying the market rate data of the noble metal futures, using reference demand information regarding the demand for the noble metal at the time of past transactions involving the noble metal and market rate data for the past noble metal future transactions, along with an association degree of three or more levels, by prioritizing the higher association degree among three or more levels between the reference demand information corresponding to the demand information acquired in the information acquisition step and the market rate data for noble metal futures.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a futures trading information display program that displays information regarding increases and decreases in futures being traded. [Background technology]

[0002] In recent years, applications that provide various advice on futures trading and systems that automate futures trading, i.e., automated trading, have become popular. When using such applications and systems, advice based on analysis of large amounts of data can increase the win rate and profits. However, even if such a large amount of data can be obtained, analyzing it and compiling an output solution that can provide accurate advice to the user requires considerable effort. Furthermore, no systems capable of performing these tasks automatically have been proposed to date. Summary of the Invention [Problem to be solved by the invention]

[0003] The present invention has been devised in consideration of the above-mentioned problems, and its purpose is to provide a futures trading information display program that displays information regarding increases and decreases in futures being traded. [Means for solving the problem]

[0004] In order to solve the above-mentioned problems, a futures trading information display program to which the present invention is applied is a futures trading information display program that displays information regarding the market prices of precious metal futures being traded, and is characterized in that the program has a computer execute an information acquisition step of acquiring demand information regarding the demand at the time of trading of precious metals, and a display step of using a correlation of three or more levels between reference demand information regarding the demand of precious metals at the time of trading of precious metals traded in the past and the market price data of precious metal futures traded in the past, and giving priority to reference demand information corresponding to the demand information acquired in the information acquisition step and market price data of precious metal futures that has a higher correlation of three or more levels. [Effects of the Invention]

[0005] Even if you do not have any particular skills or experience, you will be able to provide useful advice on futures trading. [Brief explanation of the drawings]

[0006] [Figure 1] 1 is a block diagram showing the overall configuration of a system to which the present invention is applied. [Figure 2] FIG. 2 is a diagram illustrating a specific configuration example of a search device. [Figure 3] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 4] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 5] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 6] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 7] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 8] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 9] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 10] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 11]FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 12] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 13] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 14] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 15] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 16] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 17] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 18] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 19] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 20] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 21] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 22] FIG. 10 is a diagram for explaining the operation of the present invention. [Figure 23] FIG. 10 is a diagram for explaining the operation of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0007] A futures trading information display program to which the present invention is applied will be described in detail below with reference to the drawings.

[0008] First embodiment 1 is a block diagram showing the overall configuration of a futures trading information display system 1 in which a futures trading information display program according to the present invention is implemented. The futures trading information display system 1 includes an information acquisition unit 9, a search device 2 connected to the information acquisition unit 9, and a database 3 connected to the search device 2.

[0009] The information acquisition unit 9 is a device through which users of the system input various commands and information. Specifically, it is composed of a keyboard, buttons, a touch panel, a mouse, switches, etc. The information acquisition unit 9 is not limited to a device for inputting text information, but may also be composed of a device capable of detecting voice and converting it into text information, such as a microphone. The information acquisition unit 9 may also be composed of an imaging device capable of capturing images, such as a camera. The information acquisition unit 9 may also be composed of a scanner capable of recognizing character strings from paper documents. The information acquisition unit 9 may also be integrated with the search device 2 described below. The information acquisition unit 9 outputs the detected information to the search device 2. The information acquisition unit 9 may also be composed of a means for identifying location information by scanning map information. The information acquisition unit 9 may also include a temperature sensor, a humidity sensor, a flow rate sensor, and other sensors capable of identifying substances and physical properties. The information acquisition unit 9 may also be composed of a means for automatically importing character strings and data posted on Internet sites.

[0010] Various information necessary for displaying futures trading information is accumulated in database 3. The information necessary for displaying futures trading information includes reference market condition information on past market conditions, reference event information reflecting events that occurred during the detection period of past market conditions, reference external environment information reflecting the external environment during the detection period of past market conditions, reference household information reflecting statistical data on households during the detection period of past market conditions, reference real estate information reflecting statistical data on real estate during the detection period of past market conditions, reference expert opinion information reflecting expert opinions published during the detection period of past market conditions, reference natural environment information reflecting information on the natural environment during the detection period of past market conditions, and data sets of increase / decrease data for each futures during those past market conditions.

[0011] In other words, in addition to this reference market information, database 3 stores one or more of reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, and reference natural environment information, as well as data on increases and decreases in each futures in past market conditions, all of which are linked to each other.

[0012] The search device 2 is configured with an electronic device such as a personal computer (PC), but may also be realized with any other electronic device other than a PC, such as a mobile phone, a smartphone, a tablet terminal, a wearable terminal, etc. The user can obtain a search solution by this search device 2.

[0013] 2 shows a specific example of the configuration of the search device 2. In this search device 2, a control unit 24 for controlling the entire search device 2, an operation unit 25 for inputting various control commands via operation buttons, a keyboard, etc., a communication unit 26 for performing wired or wireless communication, an estimation unit 27 for making various judgments, and a memory unit 28, typified by a hard disk or the like, for storing programs for performing searches to be executed, are all connected to an internal bus 21. Furthermore, a display unit 23 serving as a monitor for actually displaying information is connected to this internal bus 21.

[0014] The control unit 24 is a so-called central control unit that controls each component implemented in the searching device 2 by transmitting a control signal via the internal bus 21. The control unit 24 also transmits various control commands via the internal bus 21 in response to operations via the operation unit 25.

[0015] The operation unit 25 is embodied by a keyboard or a touch panel, and an execution command for executing a program is input by the user. When the execution command is input by the user, the operation unit 25 notifies the control unit 24. Upon receiving this notification, the control unit 24 executes the desired processing operation in cooperation with the estimation unit 27 and other components. The operation unit 25 may be embodied as the information acquisition unit 9 described above.

[0016] The estimation unit 27 estimates a search solution. When performing the estimation operation, the estimation unit 27 reads out various pieces of information stored in the storage unit 28 and various pieces of information stored in the database 3 as necessary information. The estimation unit 27 may be controlled by artificial intelligence. This artificial intelligence may be based on any well-known artificial intelligence technology.

[0017] The display unit 23 is configured by a graphic controller that creates a display image under the control of the control unit 24. The display unit 23 is realized by, for example, a liquid crystal display (LCD) or the like.

[0018] When the storage unit 28 is configured as a hard disk, predetermined information is written to each address and read out as necessary under the control of the control unit 24. The storage unit 28 also stores a program for carrying out the present invention. The program is read out and executed by the control unit 24.

[0019] The operation of the futures trading information display system 1 configured as described above will now be described.

[0020] The futures trading information display system 1 is used in futures trading and is premised on the fact that three or more levels of correlation between reference market information and the increase / decrease data of each futures are pre-set and acquired, as shown in FIG. 3 . The reference market information is various information related to market conditions. Examples of this reference market information include price movements of interest rates, futures, foreign exchange, stock prices of each stock, crude oil, precious metals, Bitcoin, etc. This reference market information may be displayed as a time-series chart or line graph for these items. Information such as Bollinger bands, trading volume, MACD, moving averages, etc. may also be added to this market information. This market information may also include charts of each futures and stock, Bollinger bands, MACD, moving averages, etc. For futures, information such as charts showing price movements between each futures, Bollinger bands, MACD, moving averages, etc. may also be added. This reference market information is acquired before an actual increase / decrease in futures is predicted.

[0021] The term "futures" here refers to all futures that can be traded, including agricultural products such as soybeans and corn, oil, gold, precious metals, rubber, seafood, and intangible stock price indices.

[0022] The increase / decrease data for each futures indicates the increase / decrease of each futures at a point in time after the reference market information was acquired. This increase / decrease data may be counted as the actual increase / decrease range or expressed as an increase / decrease rate. This increase / decrease data represents the increase / decrease of the futures at the measurement point (later point in time) compared to the futures at the previous point in time (i.e., the point in time when the reference market information was acquired). The previous point in time may be any time interval relative to the measurement point, such as 10 seconds, 1 minute, 30 minutes, 1 hour, 4 hours, 1 day, 10 days, 1 month, 1 year, or 5 years ago. In other words, when a certain point in time on the chart is considered the measurement point, the increase / decrease data for the futures indicates the increase / decrease of the futures at that measurement point compared to the futures at the previous point in time. Alternatively, this futures increase / decrease data may represent the actual bar on a futures chart.

[0023] In other words, through this reference market information and the dataset of futures increase / decrease data, it is possible to see how futures increased or decreased after various technical events occurred in the reference market information (for example, when the chart goes up for three consecutive days, or when an upper wick that temporarily reached a high price appears on the chart). In other words, the dataset is the increase / decrease results of futures in response to technical events. Therefore, by collecting the dataset of reference market information and futures increase / decrease data, it is possible to see how futures increased or decreased after certain market conditions in the past.

[0024] In the example of Figure 3, the input data is assumed to be, for example, reference market information P01 to P03. Such reference market information as input data is linked to the output. In this output, the increase / decrease data of futures is displayed as the output solution.

[0025] The reference market information is correlated with the futures increase / decrease data as the output solution through three or more levels of correlation. The reference market information is arranged on the left side according to this correlation, and the increase / decrease data for each futures is arranged on the right side according to the correlation. The correlation indicates the degree to which futures increase / decrease data is highly related to the reference market information arranged on the left side. In other words, the correlation is an index showing which futures increase / decrease data each reference market information is likely to be linked to, and indicates the accuracy of selecting the most likely futures increase / decrease data from the reference market information. In the example of Figure 3, correlations w13 to w19 are shown. These w13 to w19 are shown on a 10-point scale as shown in Table 1 below. The closer to 10 points, the more highly correlated each combination as an intermediate node is with the futures increase / decrease data as output. Conversely, the closer to 1 point, the less correlated each combination as an intermediate node is with the futures increase / decrease data as output.

[0026] [Table 1]

[0027] The search device 2 acquires in advance three or more levels of correlation w13 to w19 as shown in Fig. 3. In other words, the search device 2 accumulates past data on which of the reference market information and the futures increase / decrease data in that case was adopted when determining the actual search solution, and analyzes and interprets this data to create the correlations shown in Fig. 3.

[0028] For example, suppose a reference market information is when a leading signal on a futures chart surpasses a lagging signal from bottom to top. In such market conditions, let's say that there are many cases in which corn prices rise by 1% at a later point in time. In such a case, the correlation of a 1% rise in that futures will be strong. In contrast, in the exact same market conditions, there are many cases in which corn prices fall by 0.5% at a later point in time, and few cases in which corn prices rise by 1%. In such a case, the correlation of a 0.5% fall in that futures will be strong, and the correlation of a 1% rise in that futures will be weak.

[0029] This analysis may be performed using artificial intelligence. In such cases, for example, in the case of reference market information P01, analysis is performed using past price movement data for each futures. This may be extracted, for example, from electronic data of past futures charts. In the case of reference market information P01, if there are many cases of increase / decrease data A1 (gold up 4%) for each futures, the correlation degree connecting to this increase / decrease data A1 is set higher, and if there are many cases of increase / decrease data A3 (corn up 5%), the correlation degree connecting to this increase / decrease data A3 is set higher. For example, in the example of reference market information P01, increase / decrease data A1 is linked to increase / decrease data A3, but the correlation degree of w13 connecting to increase / decrease data A1 from previous cases is set to 7 points, and the correlation degree of w14 connecting to increase / decrease data A3 is set to 2 points.

[0030] The correlation shown in Fig. 3 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the correlation described above. Furthermore, the correlation is not limited to a neural network, and may be configured by any decision-making factors that constitute artificial intelligence.

[0031] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to predict futures fluctuations when actually advising new clients on futures trading. In such cases, market information relating to the market conditions at the time of the actual new futures trading will be obtained. This market information will consist of the same type of data as the reference market information described above.

[0032] The newly acquired market information is input by the above-mentioned information acquisition unit 9. The information acquisition unit 9 may acquire charts, price movement data, and the like as electronic data.

[0033] Based on the newly acquired market information in this way, future futures (i.e., future futures increase / decrease data) that are likely to occur in relation to the market information are predicted. In this case, the correlation degrees shown in Figure 3 (Table 1) obtained in advance are referenced. For example, if the newly acquired market information is identical to or similar to P02, the increase / decrease data A2 is associated with correlation degree w15, and the increase / decrease data A3 is associated with correlation degree w16. In this case, the increase / decrease data A2 with the highest correlation degree is selected as the optimal solution. However, it is not essential to select the one with the highest correlation degree as the optimal solution; the increase / decrease data A3, which has a low correlation degree but is recognized as being correlated, may be selected as the optimal solution. Of course, other output solutions without connected arrows may also be selected, and any other priority order may be used as long as it is based on correlation degree.

[0034] In this way, the possible future situations of each futures can be searched for using newly acquired market information through futures increase / decrease data and displayed to the user (consultant). By viewing the search results, the user (consultant) can obtain guidance on which futures to buy and sell based on the searched futures increase / decrease data. Simply showing the search results of futures increase / decrease data can provide useful advice to the user. Incidentally, in constructing this advice, in addition to simply displaying the searched futures increase / decrease data, it is also possible to construct advice by displaying specific amounts of which futures should be purchased or sold based on this increase / decrease data.

[0035] In the example of Figure 4, the input data is assumed to be, for example, reference market information P01 to P03 and reference event information P14 to P17. The intermediate nodes shown in Figure 4 are formed by combining the reference market information as input data with the reference event information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for each future is displayed as an output solution.

[0036] The example in Figure 4 is based on the premise that a combination of reference market information and reference event information has been formed. Reference event information is a concept that includes various social news, events, incidents, celebrations, and happy occasions that have occurred domestically or internationally, as well as news, events, incidents, celebrations, and happy occasions that have occurred for individual companies. This reference event information can be obtained from blogs, analyst reports, securities reports, advertisements, press releases, news articles, and the like related to individual companies or society as a whole. This reference event information may be extracted from character strings and dependencies that are analyzed through text mining of news articles.

[0037] In the example of Figure 4, the input data is assumed to be, for example, reference market information P01 to P03 and reference event information P14 to P17. The intermediate nodes shown in Figure 4 are formed by combining the reference market information as input data with the reference event information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for each future is displayed as an output solution.

[0038] Each combination (intermediate node) of reference market information and reference event information is correlated with the increase / decrease data of each futures product as the output solution through three or more levels of correlation. The reference market information and reference event information are arranged on the left side via this correlation, and the increase / decrease data of each futures product are arranged on the right side via this correlation. The correlation indicates the degree to which the increase / decrease data of each futures product is highly related to the reference market information and reference event information arranged on the left side. In other words, this correlation is an indicator of the likelihood that each reference market information and reference event information will be linked to which increase / decrease data of each futures product, and indicates the accuracy of selecting the most likely increase / decrease data of each futures product from the reference market information and reference event information. The increase / decrease data of each futures product at a later point in time will differ depending on market data as well as various events that actually occur in society as a whole or within each company. Therefore, by combining these reference market information and reference event information, optimal futures increase / decrease data will be searched for.

[0039] In the example of Figure 4, w13 to w22 are shown as the degrees of association. These w13 to w22 are shown on a 10-point scale as shown in Table 1, with the closer to 10 points the higher the degree of association between each combination as an intermediate node and the output, and conversely, the closer to 1 point the lower the degree of association between each combination as an intermediate node and the output.

[0040] The search device 2 acquires in advance three or more levels of correlation w13 to w22 as shown in Fig. 4. In other words, the search device 2 accumulates past data to determine which of the reference market information, reference event information, and the increase / decrease data of each futures in that case was most suitable when determining the actual search solution, and analyzes and interprets these to create the correlations shown in Fig. 4.

[0041] This analysis may be performed using artificial intelligence. In such a case, for example, if the reference market information P01 is the reference event information P16, the increase / decrease data for each futures contract is analyzed from past data. If there are many instances of increase / decrease data A1 (gold up 4%) in the increase / decrease data for each futures contract, the correlation level for this increase / decrease data A1 is set higher. If there are many instances of increase / decrease data A2 (gold down 2%) and few instances of increase / decrease data A1, the correlation level for increase / decrease data A2 is set higher and the correlation level for increase / decrease data A1 is set lower. For example, in the case of intermediate node 61a, which is linked to the output of increase / decrease data A1 and increase / decrease data A2, the correlation level for w13, which connects to increase / decrease data A1 from previous cases, is set to 7 points, and the correlation level for w14, which connects to increase / decrease data A2, is set to 2 points.

[0042] The correlation shown in Figure 4 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the correlation described above. Furthermore, the correlation is not limited to a neural network, and may be configured by any decision-making factors that constitute artificial intelligence.

[0043] 4, node 61b is a node for the combination of reference market information P01 and reference event information P14, where the degree of association for increase / decrease data A3 is w15 and the degree of association for increase / decrease data A5 is w16. Node 61c is a node for the combination of reference event information P15 and P17 for reference market information P02, where the degree of association for increase / decrease data A2 is w17 and the degree of association for increase / decrease data A4 is w18.

[0044] This correlation is what is known as learned data in artificial intelligence. After creating this learned data, the above-mentioned learned data will be used when actually searching for increase / decrease data for displaying futures trading information. In such cases, market information regarding the market conditions at the time of the new futures trading is obtained, and event information reflecting events that occurred at the time of the new futures trading is obtained. This event information corresponds to the above-mentioned reference event information, and may be obtained by importing data from news, newspapers, blogs, etc., or by directly inputting it.

[0045] In this way, optimal increase / decrease data for each futures is searched for based on the newly acquired market information and event information. In such cases, the correlation degrees shown in Figure 4 (Table 1) obtained in advance are referenced. For example, if the newly acquired market information is identical to or similar to P02 and the event information is P17, node 61d is associated via correlation degrees, and this node 61d is associated with increase / decrease data A3 at a correlation degree of w19 and increase / decrease data A4 at a correlation degree of w20. In such a case, increase / decrease data A3, which has the highest correlation degree, is selected as the optimal solution. However, it is not necessary to select the one with the highest correlation degree as the optimal solution; increase / decrease data A4, which has a low correlation degree but is recognized as having correlation, may also be selected as the optimal solution. Of course, other output solutions not connected by arrows may also be selected, and any other priority order may be used as long as it is based on correlation degrees.

[0046] Table 2 below shows examples of the degrees of association w1 to w12 extending from the input.

[0047] [Table 2]

[0048] The intermediate node 61 may be selected based on the degrees of association w1 to w12 extending from this input. In other words, the greater the degrees of association w1 to w12, the heavier the weighting in selecting the intermediate node 61. However, the degrees of association w1 to w12 may all have the same value, and the weighting in selecting the intermediate node 61 may all be the same.

[0049] FIG. 5 shows an example in which three or more levels of correlation are set between the combination of the above-mentioned reference market information and reference external environment information and the increase / decrease data of each futures for the combination.

[0050] Reference external environmental information includes various data related to politics, economy, society, technology, etc. outside the company, such as GDP, employment statistics, mining and manufacturing production index, capital investment, labor force survey, economic activity index, consumer expenditure, new car sales volume, and consumer price index.

[0051] In the example of Figure 5, the input data is assumed to be, for example, reference market information P01 to P03 and reference external environment information P18 to P21. The intermediate nodes shown in Figure 5 are formed by combining the reference market information as input data with the reference external environment information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for each future is displayed as an output solution.

[0052] Each combination (intermediate node) of reference market information and reference external environment information is correlated with the increase / decrease data for each futures contract through three or more levels of correlation. The reference market information and reference external environment information are arranged on the left side via this correlation, and the increase / decrease data are arranged on the right side via this correlation. The correlation indicates the degree to which the increase / decrease data is highly related to the reference market information and reference external environment information arranged on the left side. In other words, this correlation is an indicator of the likelihood that each reference market information and reference external environment information will be linked to the increase / decrease data, and indicates the accuracy of selecting the most likely increase / decrease data for each futures contract from the reference market information and reference external environment information. Futures change depending on market data as well as the state of the actual external environment. Therefore, the optimal increase / decrease data for each futures contract is searched for by combining these reference market information and reference external environment information.

[0053] In the example of Figure 5, w13 to w22 are shown as the degrees of association. These w13 to w22 are shown on a 10-point scale as shown in Table 1, with the closer to 10 points the higher the degree of association between each combination as an intermediate node and the output, and conversely, the closer to 1 point the lower the degree of association between each combination as an intermediate node and the output.

[0054] The search device 2 acquires in advance three or more levels of correlation w13 to w22 as shown in Fig. 5. In other words, the search device 2 accumulates past data on which of the reference market information, reference external environment information, and increase / decrease data in each case was most suitable when determining the actual search solution, and analyzes and interprets this data to create the correlations shown in Fig. 5.

[0055] This analysis may be performed using artificial intelligence. In such a case, for example, when reference market information P01 is used and reference external environment information P20 is used, the increase / decrease data is analyzed from past data. For example, in the example of intermediate node 61a, links are made to the outputs of increase / decrease data A1 and increase / decrease data A2, and from previous cases, the correlation degree of w13 connected to increase / decrease data A1 is set to 7 points, and the correlation degree of w14 connected to increase / decrease data A2 is set to 2 points.

[0056] The correlation shown in Fig. 5 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the correlations described above. Furthermore, the correlations are not limited to neural networks, and may be configured by any decision-making factors that constitute artificial intelligence.

[0057] 5, node 61b is a node representing the combination of reference market information P01 and reference external environment information P18, with the degree of correlation of increase / decrease data A3 being w15 and the degree of correlation of increase / decrease data A5 being w16. Node 61c is a node representing the combination of reference market information P02 and reference external environment information P19 and P21, with the degree of correlation of increase / decrease data A2 being w17 and the degree of correlation of increase / decrease data A4 being w18.

[0058] This correlation is what is called learned data in artificial intelligence. After creating this learned data, the advice will be provided using the learned data. In such cases, in addition to the market data described above, external environment information that reflects the external environment at the time of new futures trading is acquired. For example, if the external environment information is employment statistics information, that data may be directly imported. If it is other statistical data, that data may be directly acquired.

[0059] Based on the newly acquired market information and external environment information, the increase / decrease data for each futures is searched to construct optimal advice. In this case, the correlation degrees shown in Figure 5 (Table 1) obtained in advance are referenced. For example, if the newly acquired market information is identical to or similar to P02 and the external environment information is P21, node 61d is associated via correlation degrees, and this node 61d is associated with increase / decrease data A3 at a correlation degree of w19 and increase / decrease data A4 at a correlation degree of w20. In this case, increase / decrease data A3, which has the highest correlation degree, is selected as the optimal solution. However, selecting the one with the highest correlation degree as the optimal solution is not essential; increase / decrease data A4, which has a low correlation degree but is recognized as being correlated, may be selected as the optimal solution. Of course, other output solutions without connected arrows may also be selected, and any other priority order may be used as long as it is based on correlation degrees.

[0060] FIG. 6 shows an example in which three or more levels of correlation are set between the combination of the above-mentioned reference market information and reference household information and the increase / decrease data of each futures for the combination.

[0061] The reference household information includes various data such as household consumption surveys, household data, average working hours per week, statistical data on savings, statistical data on annual income, and data related to household finances.

[0062] In the example of Figure 6, the input data is assumed to be, for example, reference market information P01 to P03 and reference household information P22 to P25. The intermediate node shown in Figure 6 is a combination of the reference market information as input data and the reference household information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for each future is displayed as an output solution.

[0063] Each combination (intermediate node) of reference market information and reference household information is correlated with the change data for each futures contract through three or more levels of correlation. The reference market information and reference household information are arranged on the left side via this correlation, and the change data are arranged on the right side via this correlation. The correlation indicates the degree to which the change data is highly related to the reference market information and reference household information arranged on the left side. In other words, this correlation is an indicator of the likelihood that each reference market information and reference household information will be linked to the change data, and indicates the accuracy of selecting the most likely change data for each futures contract from the reference market information and reference household information. In addition to market data, there are futures whose changes fluctuate depending on the actual state of household finances. Therefore, the optimal change data for each futures contract is searched for by combining these reference market information and reference household information.

[0064] In the example of Figure 6, w13 to w22 are shown as the degrees of association. These w13 to w22 are shown on a 10-point scale as shown in Table 1, with the closer to 10 points the higher the degree of association between each combination as an intermediate node and the output, and conversely, the closer to 1 point the lower the degree of association between each combination as an intermediate node and the output.

[0065] The search device 2 acquires in advance three or more levels of correlation w13 to w22 as shown in Fig. 6. In other words, when determining an actual search solution, the search device 2 accumulates past data on which of the reference market information, reference household information, and the increase / decrease data in each case was most suitable, and analyzes and interprets these to create the correlations shown in Fig. 6.

[0066] This analysis may be performed using artificial intelligence. In such a case, for example, when reference market information P01 is reference household information P24, the increase / decrease data is analyzed from past data. For example, in the example of intermediate node 61a, the output of increase / decrease data A1 and increase / decrease data A2 are linked, and from previous cases, the correlation degree of w13 connected to increase / decrease data A1 is set to 7 points, and the correlation degree of w14 connected to increase / decrease data A2 is set to 2 points.

[0067] The correlation shown in Fig. 6 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the correlations described above. Furthermore, the correlations are not limited to neural networks, and may be configured by any decision-making factors that constitute artificial intelligence.

[0068] 6, node 61b is a node that combines reference market information P01 with reference household information P22, and the degree of association for increase / decrease data A3 is w15 and the degree of association for increase / decrease data A5 is w16. Node 61c is a node that combines reference market information P02 with reference household information P23 and P25, and the degree of association for increase / decrease data A2 is w17 and the degree of association for increase / decrease data A4 is w18.

[0069] This correlation becomes what is called learned data in artificial intelligence. After creating this learned data, the advice will be provided using the learned data. In such cases, in addition to the market data described above, household information that reflects statistical data on households at the time of new futures trading is acquired. If the household information is data published by various government agencies, such as statistical data on savings amounts, that data may be directly imported. If it is other statistical data, that data may be directly acquired.

[0070] Based on the newly acquired market information and household information, the increase / decrease data for each futures is searched for to construct optimal advice. In this case, the correlation values ​​shown in Figure 6 (Table 1) obtained in advance are referenced. For example, if the newly acquired market information is identical to or similar to P02 and the household information is P25, node 61d is associated via correlation values, and this node 61d is associated with increase / decrease data A3 at a correlation value of w19 and increase / decrease data A4 at a correlation value of w20. In this case, increase / decrease data A3, which has the highest correlation value, is selected as the optimal solution. However, selecting the one with the highest correlation value as the optimal solution is not essential; increase / decrease data A4, which has a low correlation value but is recognized as being correlated, may be selected as the optimal solution. Of course, other output solutions not connected by arrows may also be selected, and any other priority may be selected based on correlation values.

[0071] Note that reference real estate information that reflects statistical data on real estate at the time of detecting past market conditions may be used as input data instead of the reference household information shown in Figure 6. Detailed explanation of the configuration in such a case will be omitted by replacing reference household information with reference real estate information and household information with real estate information.

[0072] In such a case, three or more levels of correlation will be used between the combination of reference market information and reference real estate information and the increase / decrease data of each of the above futures. Node 61 will specify three or more levels of correlation between the combination of reference market information and reference real estate information and the increase / decrease data of each of the above futures.

[0073] Reference real estate information includes all information related to real estate, such as office vacancy rates, unit prices per square meter, average rents, land prices, and statistical data on vacant homes.

[0074] In providing advice on futures trading, after forming such correlations, real estate information that reflects statistical data on real estate at the time of new futures trading is obtained, and advice on futures trading is provided based on a combination of the same or similar reference real estate information and reference market information, and three or more levels of correlation with the increase / decrease data of each futures.

[0075] FIG. 7 shows an example in which three or more levels of correlation are set between the combination of the above-mentioned reference market information and reference expert opinion information and the increase / decrease data of each futures for that combination.

[0076] Reference expert opinion information refers to any information that expresses expert opinions regarding the increase or decrease of futures, such as futures forecasts published in analyst reports, newspaper articles, etc., or expert comments and opinions regarding the reasons for stock price increases or decreases. Reference expert opinion information may also simply be predictions regarding whether each future will rise, fall, or remain unchanged. This reference expert opinion information may include opinions regarding Nikkei 225 futures as a whole, opinions regarding a specific segment or industry, or even opinions regarding individual futures. Reference expert opinion information may also include comments and rise or fall forecasts by experts (analysts) published on the Internet.

[0077] In the example of Figure 7, the input data is assumed to be, for example, reference market information P01 to P03 and reference expert opinion information P26 to P29. The intermediate nodes shown in Figure 7 are formed by combining the reference market information as input data with the reference expert opinion information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for each future is displayed as an output solution.

[0078] Each combination (intermediate node) of reference market information and reference expert opinion information is correlated with the increase / decrease data for each futures contract through three or more levels of correlation. The reference market information and reference expert opinion information are arranged on the left side via correlation, and the increase / decrease data are arranged on the right side via correlation. Correlation indicates the degree to which the increase / decrease data is highly related to the reference market information and reference expert opinion information arranged on the left side. In other words, this correlation is an indicator of the likelihood that each reference market information and reference expert opinion information is linked to the increase / decrease data, and indicates the accuracy of selecting the most likely increase / decrease data for each futures contract from the reference market information and reference expert opinion information. Futures fluctuations may be correlated with not only market data but also actual expert opinions. Therefore, the optimal increase / decrease data for each futures contract is searched for by combining these reference market information and reference expert opinion information.

[0079] In the example of Figure 7, w13 to w22 are shown as the degrees of association. These w13 to w22 are shown on a 10-point scale as shown in Table 1, with the closer to 10 points the higher the degree of association between each combination as an intermediate node and the output, and conversely, the closer to 1 point the lower the degree of association between each combination as an intermediate node and the output.

[0080] The search device 2 acquires in advance three or more levels of correlation w13 to w22 as shown in Fig. 7. In other words, when determining an actual search solution, the search device 2 accumulates past data on which of the reference market information, reference expert opinion information, and increase / decrease data in each case was most suitable, and analyzes and interprets these to create the correlations shown in Fig. 7.

[0081] This analysis may be performed using artificial intelligence. In such a case, for example, when reference market information P01 is reference household information P28, the increase / decrease data is analyzed from past data. For example, in the example of intermediate node 61a, the output of increase / decrease data A1 and increase / decrease data A2 are linked, and from previous cases, the correlation degree of w13 connected to increase / decrease data A1 is set to 7 points, and the correlation degree of w14 connected to increase / decrease data A2 is set to 2 points.

[0082] 7 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the above-mentioned degrees of association. Furthermore, the degrees of association are not limited to neural networks, and may be configured by any decision-making factors that constitute artificial intelligence.

[0083] 7, node 61b is a node that combines reference market information P01 with reference expert opinion information P26, where the degree of association for increase / decrease data A3 is w15 and the degree of association for increase / decrease data A5 is w16. Node 61c is a node that combines reference market information P02 with reference expert opinion information P27 and P29, where the degree of association for increase / decrease data A2 is w17 and the degree of association for increase / decrease data A4 is w18.

[0084] This correlation becomes what is called learned data in artificial intelligence. After creating this learned data, the advice will be provided using the learned data. In such cases, in addition to the market data described above, expert opinion information that reflects the opinions of experts published at the time of new futures trading is acquired. For example, if an expert's opinion is expressed in a newspaper article, that data may be directly imported as the expert opinion information.

[0085] Based on the newly acquired market information and expert opinion information, the increase / decrease data for each futures is searched to construct optimal advice. In this case, the correlation values ​​shown in Figure 7 (Table 1) obtained in advance are referenced. For example, if the newly acquired market information is identical to or similar to P02 and the expert opinion information is P29, node 61d is associated via correlation values, and this node 61d is associated with increase / decrease data A3 at a correlation value of w19 and increase / decrease data A4 at a correlation value of w20. In this case, increase / decrease data A3, which has the highest correlation value, is selected as the optimal solution. However, selecting the one with the highest correlation value as the optimal solution is not essential; increase / decrease data A4, which has a low correlation value but is recognized as being correlated, may be selected as the optimal solution. Of course, output solutions without arrows may also be selected, and any other priority may be selected based on correlation values.

[0086] FIG. 8 shows an example in which three or more levels of correlation are set between the combination of the above-mentioned reference market information and reference natural environment information and the increase / decrease data of each futures for the combination.

[0087] Reference natural environment information refers to all information related to the natural environment, such as disaster data, temperature data, precipitation data, wind direction data, humidity data, etc., and includes data on the past natural environment published by the Japan Meteorological Agency or data on the past natural environment published by private companies or individuals.

[0088] In the example of Figure 8, the input data is assumed to be, for example, reference market information P01 to P03 and reference natural environment information P30 to P33. The intermediate nodes shown in Figure 8 are formed by combining the reference market information as input data with the reference natural environment information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for each future is displayed as an output solution.

[0089] Each combination (intermediate node) of reference market information and reference natural environment information is correlated with the fluctuation data of each futures product in the output solution through three or more levels of correlation. The reference market information and reference natural environment information are arranged on the left side via this correlation, and the fluctuation data are arranged on the right side via this correlation. The correlation indicates the degree to which the fluctuation data is highly related to the reference market information and reference natural environment information arranged on the left side. In other words, this correlation is an indicator of the likelihood that each reference market information and reference natural environment information will be linked to the fluctuation data, and it indicates the accuracy of selecting the most likely fluctuation data for each futures product from the reference market information and reference natural environment information. Futures fluctuations may be correlated with market data as well as actual expert opinions. Therefore, the optimal fluctuation data for each futures product is searched for by combining these reference market information and reference natural environment information.

[0090] In the example of Figure 8, w13 to w22 are shown as the degrees of association. These w13 to w22 are shown on a 10-point scale as shown in Table 1, with the closer to 10 points the higher the degree of association between each combination as an intermediate node and the output, and conversely, the closer to 1 point the lower the degree of association between each combination as an intermediate node and the output.

[0091] The search device 2 acquires in advance three or more levels of correlation w13 to w22 as shown in Fig. 8. In other words, when determining an actual search solution, the search device 2 accumulates past data on which of the reference market information, reference natural environment information, and increase / decrease data in each case was most suitable, and analyzes and interprets these to create the correlations shown in Fig. 8.

[0092] This analysis may be performed using artificial intelligence. In such a case, for example, when reference market information P01 and reference natural environment information P32 are used, the increase / decrease data is analyzed from past data. For example, in the case of intermediate node 61a to which reference market information P01 and reference natural environment information P32 are linked, the output of increase / decrease data A1 and increase / decrease data A2 are linked, and the correlation degree of w13 connected to increase / decrease data A1 is set to 7 points from previous cases, and the correlation degree of w14 connected to increase / decrease data A2 is set to 2 points.

[0093] The correlation shown in Fig. 8 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the correlation described above. Furthermore, the correlation is not limited to a neural network, and may be configured by any decision-making factors that constitute artificial intelligence.

[0094] 8, node 61b is a node that combines reference market information P01 with reference natural environment information P30, and the degree of association for increase / decrease data A3 is w15 and the degree of association for increase / decrease data A5 is w16. Node 61c is a node that combines reference market information P02 with reference expert opinion information P31 and P33, and the degree of association for increase / decrease data A2 is w17 and the degree of association for increase / decrease data A4 is w18.

[0095] This correlation is what is known as learned data in artificial intelligence. After creating this learned data, the advice will be provided using the learned data. In such cases, in addition to the market data described above, natural environment information that reflects the natural environment information at the time of new futures trading is obtained. Natural environment information may be obtained, for example, from data and information on the natural environment published by the Japan Meteorological Agency, private companies, or individuals, or directly from websites that contain such information.

[0096] Based on the newly acquired market condition information and natural environment information, the increase / decrease data for each futures is searched to generate optimal advice. In this case, the correlations shown in Figure 8 (Table 1) obtained in advance are referenced. For example, if the newly acquired market condition information is identical to or similar to P02 and the natural environment information is P33, node 61d is associated via correlation, and this node 61d is associated with increase / decrease data A3 at a correlation level of w19 and increase / decrease data A4 at a correlation level of w20. In this case, increase / decrease data A3, which has the highest correlation, is selected as the optimal solution. However, selecting the one with the highest correlation as the optimal solution is not essential; increase / decrease data A4, which has a low correlation but is recognized as being correlated, may be selected as the optimal solution. Of course, other output solutions without connected arrows may also be selected, and any other priority order may be used as long as they are based on correlation.

[0097] Figure 9 shows an example in which, in addition to the above-mentioned reference market information and reference event information, a combination of reference external environment information is also set, and three or more levels of correlation are set between the increase / decrease data for each futures for that combination.

[0098] In such a case, the correlation degree is expressed as a set of combinations of reference market information, reference event information, and reference external environment information as intermediate nodes 61a to 61e, as described above, as shown in Figure 9.

[0099] 9, node 61c is associated with reference market information P02 at a correlation level of w3, reference event information P15 at a correlation level of w7, and reference external environment information P19 at a correlation level of w11. Similarly, node 61e is associated with reference market information P03 at a correlation level of w5, reference event information P15 at a correlation level of w8, and reference external environment information P18 at a correlation level of w10.

[0100] Similarly, when such correlations are set, a search solution is determined based on newly acquired market information, event information, and external environment information.

[0101] In determining this search solution, the correlation degrees shown in Figure 9, which have been acquired in advance, are referenced. For example, if the acquired market information is identical to or similar to reference market information P02, the acquired event information corresponds to reference event information P15, and the acquired external environment information corresponds to reference external environment information P19, then that combination is associated with node 61c, and this node 61c is associated with increase / decrease data A2 at correlation degree w17 and increase / decrease data A4 at correlation degree w18. Based on these correlation degrees, w17 and w18, the search solution is actually determined.

[0102] When combining three or more types of such input parameters, it is also possible to apply a combination consisting of two or more of reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, and reference natural environment information.

[0103] Furthermore, as output data, in addition to the increase / decrease data for each futures, advice regarding actual futures purchasing behavior (e.g., buy XX futures, hold XX futures) may be directly displayed, as shown in FIG. 10. This advice may not only specify the futures for which buying / selling is recommended, but may also include specific purchase amounts. Such advice may be generated based on the increase / decrease data described above. In such cases, advice may be given to buy if the futures will be higher in the future, or to sell if the futures will be lower in the future. Furthermore, advice may also display risks in addition to the potential return of futures trading. In this case, as input data and data to be learned, it is of course possible to directly include this advice content in the dataset instead of the increase / decrease data and use it for learning.

[0104] The present invention may also be embodied as an automated futures trading program that automatically trades futures. In such a case, after searching for increase / decrease data based on the procedure described above, each future is automatically bought and sold based on the increase / decrease data. In such a case, the system itself buys and sells futures based on advice regarding futures purchasing behavior (e.g., buy futures XX, hold futures XX). In such a case, if the search for increase / decrease data for each future determines that a certain future will rise, the system automatically purchases that future. On the other hand, if the search determines that a certain future will fall, the system automatically sells or short-sells that future.

[0105] The above-mentioned correlation may be configured by a neural network node in artificial intelligence as shown in Figure 11. In other words, the weighting coefficient for the output of this neural network node corresponds to the above-mentioned correlation. Furthermore, it is not limited to a neural network, and may be configured by any decision-making factor that constitutes artificial intelligence.

[0106] Furthermore, the present invention determines the increase / decrease data for each futures based on the correlation between a combination of two or more types of information, namely, reference information U and reference information V, as shown in Fig. 12. This reference information Y is reference market information, and reference information V is any other reference information (for example, reference event information, reference expert opinion information, etc.).

[0107] 12, the output obtained for the reference information U may be used as input data as it is, and may be associated with the output (increase / decrease data for each future) via an intermediate node 61 in combination with the reference information V. For example, after an output solution is obtained for the reference information U, this may be used as input as it is, and the degree of correlation with other reference information V may be used to search for the output (increase / decrease data for each future).

[0108] Furthermore, in the present invention, charts of market information and reference market information may be applied to chart patterns for buy and sell signals. Figures 13 and 14 show examples of chart patterns for buy and sell signals. For example, in Figure 13(a), when stock prices or futures values ​​repeatedly fluctuate based on a moving average line, a buy signal occurs when the stock price or futures value falls to the moving average line. In Figure 13(b), a buy signal occurs when the stock price breaks above the upper resistance line after a prolonged period of sideways movement. Figure 13(c) is known as a double bottom pattern, in which a buy signal occurs when the stock price hits two lows in the low price range. Figure 13(d) is known as an inverse head and shoulders pattern, in which the stock price hits three lows in the low price range, with the middle one being the lowest. When this pattern appears, a buy signal occurs. Figure 14(a) shows a situation where stock prices soar, then immediately fall sharply, reversing with a long lower wick or a large bullish candlestick. When this occurs, it is a buy signal. Figure 14(b) shows what's called the "Mikawa Morning Star." A large bearish candlestick appears at the bottom, followed by a lower window opening and a bullish / bearish candlestick (koma) with short wicks and a short body. A large upper window opening and a large bullish candlestick form is a buy signal. Figure 14(c) shows what's called the "Mikawa Morning Crow." It shows a V-shaped reversal with three black soldiers (three crows) followed by three red soldiers, which is a buy signal. Figure 14(d) shows what's called the "Mikawa Evening Star." A large bullish candlestick appears during an uptrend, followed by an upper window opening and a bullish / bearish candlestick (koma) with short wicks and a short body. A large lower window opening and a large bearish candlestick form is a sell signal.

[0109] Such signals have been developed based on past experience in not only futures trading but also stock trading, but in the present invention, this market information and reference market information may be applied to the chart pattern types of these buying and selling signals.

[0110] This fitting may utilize a decision model generated by machine learning, such as that shown in FIG. 15. In this decision model, images of chart patterns of buy / sell signals, such as the examples described above, are used as training data. The input is the chart of each futures, and the output is the type of buy / sell signal. When a chart is acquired, fitting is performed based on this decision model generated by machine learning, and it is determined which type of buy / sell signal it corresponds to.

[0111] For example, when a chart is input, it can be determined whether the output corresponds to a trading signal type such as Mikawa Akegarasu or Mikawa Yoi no Myojo.

[0112] In the present invention, when machine learning is performed based on the correlation between reference market information and the increase / decrease data of each futures, the reference market information is acquired by acquiring a chart of each futures that represents the market condition. The acquired chart is then used to determine which type of trading signal it corresponds to using the determination model shown in Figure 15, and each piece of reference market information is assigned to a trading signal type.

[0113] As a result, this reference market information is expressed as categorized buy / sell signals. The correlation described above is formed by learning the increase / decrease trends of futures at a subsequent point in time in response to such buy / sell signals.

[0114] Next, even when market information is actually acquired, the acquired chart is judged to which buy / sell signal type it corresponds to using the judgment model shown in Figure 15, and each reference market information is assigned to a buy / sell signal type. As a result, this market information is represented as a categorized buy / sell signal. The type of buy / sell signal in such market information is judged to correspond to which buy / sell signal type in the reference market information using the correlation level described above. Then, using three or more levels of correlation between the buy / sell signal type in the reference market information that corresponds to the buy / sell signal type in the market information and the increase / decrease data for each futures, the increase / decrease data for each futures is displayed with a higher correlation level as the priority.

[0115] The types of buy / sell signals corresponding to the output shown in FIG. 15 are not limited to those that have already been proposed, and new signal types may be updated sequentially.

[0116] For example, if an analysis of each chart in the reference market information or market information does not find a fit with the existing types in the output shown in Figure 15, the trend of that chart is registered as a new type. Then, a correlation is formed between this newly registered type and the increase / decrease data of each futures at a subsequent point in time. If a chart similar to this newly registered type is subsequently input, a correlation is similarly formed with the increase / decrease data of each futures, and a correlation weight w is formed between this newly registered type and the increase / decrease data of each futures.

[0117] By updating this correlation, when new market information is input and the judgment model of Figure 15 determines that the input is of a newly registered signal type, it becomes possible to search for a search solution using the reference market information consisting of the newly registered type.

[0118] In the above-mentioned correlation degree, the correlation degree is expressed on a 10-point scale, but it is not limited to this and may be expressed on a scale of 3 or more, and conversely, if it is 3 or more, it may be expressed on a scale of 100 or 1000. On the other hand, this correlation degree does not include a 2-point scale, that is, a scale expressed by either 1 or 0, indicating whether or not there is a correlation between the two.

[0119] According to the present invention having the above-described configuration, anyone can easily search for the optimal futures for futures trading, even without special skills or experience. Furthermore, according to the present invention, it is possible to judge the search solution with higher accuracy than a human. Furthermore, by configuring the above-mentioned correlation using artificial intelligence (such as a neural network) and having it learn, it is possible to further improve the accuracy of the judgment.

[0120] In addition, since there are many cases where the above-mentioned input data and output data do not exist exactly the same during the learning process, the input data and output data may be classified by type. In other words, the information P01, P02, ..., P15, 16, ... that constitutes the input data may be classified according to classification criteria previously determined by the system or user depending on the content of the information, and a data set may be created using the classified input data and output data, and learning may be performed.

[0121] Furthermore, the present invention is characterized in that an optimal solution is searched for through correlation levels set to three or more levels. The correlation level can be expressed, for example, by a numerical value from 0 to 100%, in addition to the above-mentioned 10 levels, but is not limited to this and may be configured in any level as long as it can be expressed by a numerical value of three or more levels.

[0122] By determining futures with higher profit margins and lower risks based on the correlation level expressed as a numerical value in three or more levels, it is possible to search and display futures in descending order of correlation level in situations where multiple possible search solutions are considered.

[0123] In addition, according to the present invention, it is possible to judge without overlooking even a discrimination result with an extremely low output, such as a correlation degree of 1%, and it is possible to alert the user that even a discrimination result with an extremely low correlation degree is connected as a slight sign, and that it may be useful as a discrimination result once in tens or hundreds of times.

[0124] Furthermore, according to the present invention, by performing a search based on such three or more levels of correlation, there is an advantage in that the search policy can be determined by how the threshold is set. A low threshold can detect even cases with a correlation of 1% without omission, but the possibility of detecting a more appropriate discrimination result is low and a lot of noise may be picked up. On the other hand, a high threshold can detect the optimal search solution with a high probability, but it may miss a suitable solution that usually has a low correlation and is ignored, but appears once in tens or hundreds of times. The emphasis can be decided based on the user's or system's perspective, and it is possible to increase the degree of freedom in selecting the points to be emphasized.

[0125] Furthermore, in the present invention, the correlation degree may be updated. This update may reflect information provided via a public communication network such as the Internet. Furthermore, when knowledge, information, and data related to event information, external environment information, household information, real estate information, expert opinion information, and natural environment information are acquired in addition to market information, the correlation degree may be increased or decreased accordingly. Similarly, when the above-mentioned expert opinion information, natural environment information, fundamental information, and statistical information are acquired in addition to event information as a substitute for the external environment information, the correlation degree may be increased or decreased accordingly.

[0126] In other words, this update corresponds to learning in artificial intelligence. Because new data is acquired and reflected in the learned data, it can be considered a learning process.

[0127] Furthermore, the degree of association may be updated not only based on information obtainable from public communication networks, but also manually or automatically by the system or user based on the contents of research data and papers by experts, academic presentations, newspaper articles, books, etc. Artificial intelligence may be utilized in these update processes.

[0128] Furthermore, the process of initially creating a trained model and the above-mentioned updates may use not only supervised learning, but also unsupervised learning, deep learning, reinforcement learning, etc. In the case of unsupervised learning, instead of reading and learning a data set of input data and output data, information corresponding to the input data may be read and learned, and then the correlation related to the output data may be self-formed from the information.

[0129] Second embodiment In a second embodiment, the present invention provides a futures trading information display program that displays information related to the market prices of agricultural crop futures. In this second embodiment, the market prices of agricultural crop futures are estimated from image information of the growing process of agricultural crops such as soybeans, adzuki beans, and corn. For example, as shown in Fig. 16, reference image information extracted from images of previously grown agricultural crops, showing the growth status of the crops, damage caused by pests, or diseases that have occurred on the crops, is used to obtain three or more levels of correlation between the reference image information and market price data of the agricultural crop futures for the previously grown crops that have been harvested.

[0130] Reference image information refers to images of agricultural crops in the growing process captured by a camera. Images of agricultural crops may be captured sequentially in chronological order, from sowing seeds and seedlings to harvesting. Images of agricultural crops may be composed of images capturing the entire field or paddy field, or may include close-up images of the leaves, stems, and fruits of agricultural crops in the field or paddy field. Images may be captured by an unmanned aerial vehicle such as a drone, or by a ground-based camera. Such cameras can detect the growth status of agricultural crops, pest damage to agricultural crops, diseases, and the like. The reference image information may be obtained by directly capturing raw images captured by a camera, or by extracting only characteristic parts of the image using well-known deep learning technology.

[0131] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, it will be used to estimate new market prices for agricultural products. In such cases, new image information will be acquired.

[0132] The newly acquired image information is obtained by taking an image using the camera of the information acquisition unit 9. This image capture is of a crop that is about to be newly grown.

[0133] Based on the newly acquired image information in this way, the market price of agricultural products is predicted. In such cases, the correlation degree shown in Fig. 16, which has been acquired in advance, is referenced. The specific method for estimating the market price of agricultural products is the same as in the first embodiment, and therefore a detailed description thereof will be omitted below.

[0134] The yield and quality of agricultural crops vary depending on their growth conditions and the damage caused by pests, and trading prices change accordingly. Therefore, by training a dataset that shows the relationship between such reference image information and actual trading prices of agricultural crops, it becomes possible to estimate trading prices according to the growth conditions and pest damage of the crops shown in the image.

[0135] The example in FIG. 17 is based on the premise that a combination of reference image information and reference soil information has been formed. The reference soil information includes all information related to the soil where crops are grown. Examples of this reference soil information include soil components, pH, water content, temperature, etc. The results of actually collecting soil components and analyzing them using chemical analysis methods may be used, or data detected by a well-known soil sensor may be used. Images of soil captured by a camera, and then using well-known deep learning technology to extract only characteristic parts of the image, may also be used.

[0136] In the example of Figure 17, the input data is assumed to be, for example, reference image information P11 to P13 and reference soil information P14 to P17. The intermediate nodes shown in Figure 17 are formed by combining the reference image information as input data with the reference soil information. Each intermediate node is further connected to an output. In this output, the increase / decrease data of agricultural crop futures is displayed as an output solution.

[0137] Each combination (intermediate node) of reference image information and reference soil information is correlated with the agricultural product futures fluctuation data or market price as the output solution through three or more levels of correlation. The reference image information and reference soil information are arranged on the left side via this correlation, and each agricultural product futures fluctuation data or market price is arranged on the right side via this correlation. The correlation indicates the degree to which agricultural product futures fluctuation data or market price is highly related to the reference image information and reference soil information arranged on the left side. In other words, this correlation is an index indicating the likelihood that each reference image information and reference soil information is linked to which agricultural product futures fluctuation data or market price, and indicates the accuracy of selecting the most likely agricultural product futures fluctuation data or market price from the reference image information and reference soil information. In the example of Figure 17, correlations w13 to w22 are shown.

[0138] The estimation device 2 acquires in advance three or more stages of correlation w13 to w22 as shown in Fig. 17. That is, when determining an actual search solution, the estimation device 2 accumulates reference image information, reference soil information, and past data on the fluctuation data or market price of agricultural produce futures at that time, and creates the correlations shown in Fig. 16 by analyzing these.

[0139] For example, if the reference image information P11 shows the condition of agricultural crops damaged by pests, and the soil composition where the crops were actually grown is investigated and found to be "pH ●●, composition ○×" corresponding to the reference soil information P14, then previous data can be tracked to investigate the actual crops up to harvest and extract the futures increase / decrease data or market price.

[0140] The correlations shown in Fig. 17 may be configured by nodes of a neural network in artificial intelligence. That is, the weighting coefficients for the outputs of the nodes of this neural network correspond to the correlations described above. Furthermore, the correlations are not limited to neural networks, and may be configured by any decision-making factors that constitute artificial intelligence.

[0141] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, when actually determining new increase / decrease data or market prices (market price data) for agricultural crop futures, the trained data described above will be used to determine the increase / decrease data or market price for agricultural crop futures. In such cases, new image information and soil information will be acquired.

[0142] The soil information to be acquired is the soil in which the crops to be newly grown are actually planted, and the method of acquisition is the same as that for acquiring the reference soil information described above.

[0143] Based on the newly acquired image information and soil information in this way, the fluctuation data or market price of the futures of the agricultural product for which the newly acquired image information and soil information has actually been obtained is predicted. In such cases, the correlation degrees shown in FIG. 17 (Table 1) acquired in advance are referenced, and the one with the highest correlation degree is selected as the optimal solution. The details of the method for selecting this optimal solution are the same as those described in the first embodiment above.

[0144] In the second embodiment, the reference image information to be combined with the reference natural environment information may be used instead of the reference soil information to learn reference natural environment information that detects weather and disaster conditions during the growing period of the crops for which the reference image information was captured. The reference natural environment information is any data related to the natural environment detected during the growing process of the crops, such as data related to weather conditions such as solar radiation, temperature, humidity, wind direction, and rainfall, and data related to disaster conditions such as typhoons, floods, droughts, and droughts. While it is preferable to obtain this reference natural environment information as of the time the reference image information is acquired, this is not limitative. Data related to the natural environment throughout the growing period of the crops for which the reference image information was captured may also be obtained. This reference natural environment information may be composed of sensing means for obtaining real-time data, such as temperature sensors, humidity sensors, light sensors, wind vanes, and rain gauges, or may be used to analyze conditions such as damage caused by typhoons and floods, droughts, and droughts after the fact.

[0145] A correlation of three or more levels between a combination of reference image information and reference natural environment information and the market price data of the agricultural crop futures is learned. Then, new image information and natural environment information are acquired. This natural environment information is detected weather and disaster conditions during the growing season of the agricultural crops for which the image information was captured, and the type of information is the same as that of the reference natural environment information. A search solution for the market price data of the agricultural crop futures is then searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference natural environment information corresponding to the natural environment information, with a higher correlation of three or more levels between the market price data of the agricultural crop futures.

[0146] As reference information to be combined with the reference image information, reference history information regarding the history of agricultural work actually performed on previously grown crops may be learned as an alternative to reference soil information. The reference history information is a history of agricultural work actually performed during the growing process of the crop from which the reference image information was captured. It organizes the specific agricultural work performed from sowing seeds to planting seedlings and harvesting. For example, this reference history information reflects when, to what extent, and how agricultural work, such as watering, fertilizing, spraying pesticides, and weed removal, was performed. In practice, this reference history information may be composed of digitalized agricultural work logs kept by the farmer, or may be information obtained via a PC, smartphone, or the like that contains records of actual agricultural work.

[0147] A correlation of three or more levels between a combination of reference image information and reference history information and market price data for agricultural produce futures is learned. Then, new image information and history information are acquired. This history information is a history of the agricultural work actually performed in the growing process of the agricultural produce for which the image information was captured, and the type of information is the same as that of the reference history information. Then, a search solution for market price data for agricultural produce futures is searched for, giving priority to combinations having reference image information corresponding to the acquired image information and reference history information corresponding to the history information, which have a higher correlation of three or more levels between the market price data for agricultural produce futures.

[0148] As reference information to be combined with the reference image information, reference market information may be learned instead of reference soil information. The reference market information is the same as that described in the first embodiment, but the market conditions acquired as data may be any period from the growing process of the crop when the reference image information was captured until the time of trading.

[0149] In such cases, the system learns a correlation of three or more levels between a combination of reference image information and reference market information and the market price data of agricultural product futures. Then, new image information and market information are acquired. This market information may be of the same type as the reference market information, as long as it covers the period from the stage of growing the agricultural product when the image information was captured to the time of trading. Then, a search solution for agricultural product futures market price data is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference market information corresponding to the market information, which have a higher correlation of three or more levels between the combination and the market price data of agricultural product futures.

[0150] As reference information to be combined with reference image information, reference event information may be learned instead of reference soil information. The reference event information is the same as that described in the first embodiment, but the events acquired as data may be any events that occurred during the period from the growing process of the crop when the reference image information was captured until the time of transaction.

[0151] In such cases, the system learns three or more levels of correlation between combinations of reference image information and reference event information and market price data for agricultural product futures. Then, new image information and event information are acquired. This event information may be information that occurred during the period from the stage of growing the agricultural product when the image information was captured until the time of trading, and the type of information is the same as that of the reference event information. Then, a search solution for market price data for agricultural product futures is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference event information corresponding to the event information, with a higher correlation of three or more levels between the market price data for agricultural product futures.

[0152] As reference information to be combined with the reference image information, reference external environment information may be learned instead of reference soil information. The reference external environment information is the same as that described in the first embodiment, but the external environment acquired as data may be any information covering the period from the crop growing process when the reference image information was captured to the time of transaction.

[0153] In such cases, the system learns three or more levels of correlation between combinations of reference image information and reference external environment information and agricultural futures market price data. Then, new image information and external environment information are acquired. This external environment information may be from the period from the crop growth process when the image information was captured to the time of trading, and the type of information is the same as that of the reference external environment information. Then, a search solution for agricultural futures market price data is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference external environment information corresponding to the external environment information, with a higher correlation of three or more levels between the combination and agricultural futures market price data.

[0154] As reference information to be combined with reference image information, reference expert opinion information may be learned instead of reference soil information. The reference expert opinion information is the same as that described in the first embodiment, but the expert opinion acquired as data may be any opinion issued during the period from the growing process of the crop when the reference image information was captured until the time of transaction.

[0155] In such cases, the system learns a correlation of three or more levels between a combination of reference image information and reference expert opinion information and market price data for agricultural product futures. Then, new image information and expert opinion information are acquired. This expert opinion information may be generated during the period from the stage of growing the agricultural product when the image information was captured until the time of trading, and the type of information is the same as that of the reference expert opinion information. Then, a search solution for agricultural product futures market price data is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference expert opinion information corresponding to the expert opinion information, with a correlation of three or more levels between the combination and market price data for agricultural product futures.

[0156] As reference information to be combined with reference image information, reference planted area information may be learned as an alternative to reference soil information. The reference planted area information is data relating to the planted area during past agricultural crop cultivation. That is, this reference planted area information is data relating to the planted area of ​​cultivated land where actual agricultural products are produced, and may reflect, for example, not only statistical data on a global or national level, but also information relating to the planted area on a regional, prefectural, municipal, or even farmland level basis. This reference planted area information changes over time, such as monthly or annually, and this is reflected as statistical data. Therefore, learning data can be obtained by acquiring such statistical data over time.

[0157] In such cases, the system learns three or more levels of correlation between a combination of reference image information and reference planted area information and market price data for agricultural produce futures. Then, new image information and planted area information are acquired. This planted area information may be the planted area at the time of growth of the agricultural produce for which the image information was captured, and the type of information is the same as that of the reference planted area information. Then, a search solution for agricultural produce futures market price data is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference planted area information corresponding to the planted area information, with a higher level of correlation of three or more levels between the combination and market price data for agricultural produce futures.

[0158] Reference policy information may be learned as reference information to be combined with reference image information, instead of reference soil information. The reference policy information is information about policies that were implemented from the time of cultivation of the agricultural crops cultivated in the past to the time of trading. The policies mainly concern agricultural production, and include any policy that affects the supply and demand relationship of agricultural crops sold domestically, such as policies to reduce cultivated land, policies to increase cultivated land, policies to actively import agricultural crops from other countries, policies regarding tariffs on agricultural crops imported from other countries, and policies regarding the liberalization of agricultural crops imported from other countries.

[0159] In such cases, the system learns a correlation of three or more levels between a combination of reference image information and reference policy information and agricultural futures market price data. Then, new image information and policy information are acquired. This policy information is information about policies that will be implemented at the time of new futures trading, and the type of information is the same as the reference planted acreage information. Then, a search solution for agricultural futures market price data is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference policy information corresponding to the policy information, with a higher correlation of three or more levels between the combination and agricultural futures market price data.

[0160] As reference information to be combined with reference image information, reference agricultural work history information may be learned as an alternative to reference soil information. The reference agricultural work history information is a history of agricultural work actually performed during the cultivation process of the crop. It organizes the specific agricultural work performed from sowing seeds to planting seedlings and harvesting. For example, this reference history information reflects when, to what extent, and how agricultural work, such as watering, fertilizing, spraying pesticides, and weed control, was performed. In practice, this reference history information may be composed of digitalized agricultural work logs kept by farmers, or may be information obtained via a PC, smartphone, etc., containing records of actual agricultural work.

[0161] In such cases, the system learns three or more levels of correlation between combinations of reference image information and reference farm work history information and market price data for agricultural crop futures. Then, new image information and farm work history information are acquired. This farm work history information is information about the farm work history during the growing process of the agricultural crop that is the subject of the new futures transaction, and the type of information is the same as that of the reference farm work history information. Then, a search solution for agricultural crop futures market price data is searched for, giving priority to combinations of reference image information corresponding to the acquired image information and reference farm work history information corresponding to the farm work history information that have a higher level of correlation of three or more levels with market price data for agricultural crop futures.

[0162] As reference information to be combined with reference image information, reference season information may be learned as an alternative to reference soil information. The reference season information is information about the season in which agricultural products grown in the past were traded, such as the month, week, season (spring, summer, autumn, winter), or annual events such as Christmas, New Year, Obon, seasonal festivals, and Setsubun. For example, the supply and demand of agricultural products changes depending on the season, such as the consumption of azuki beans increasing during New Year's and the consumption of soybeans increasing during Setsubun, and this affects market prices. For this reason, the reference season information is also used as an explanatory variable.

[0163] In such cases, a correlation of three or more levels between a combination of reference image information and reference seasonal information and market price data for agricultural produce futures is learned. Then, new image information and seasonal information are acquired. This seasonal information is information about the season in which new futures transactions will be conducted, and the type of information is the same as that of the reference seasonal information. Then, a search solution for market price data for agricultural produce futures is searched for, giving priority to combinations having reference image information corresponding to the acquired image information and reference seasonal information corresponding to the seasonal information, which have a higher correlation of three or more levels between the market price data for agricultural produce futures.

[0164] Furthermore, the first to second embodiments are not limited to the above-described embodiments. For example, as shown in FIG. 18, three or more levels of correlation between the reference information serving as the base and the futures price may be used. In such a case, a solution search is performed based on three or more levels of correlation between the reference information according to newly acquired information and the futures price. The base reference information may be, for example, reference image information, but is not limited to this. Any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference agricultural work history information, reference expert opinion information, reference natural environment information, reference soil information, reference planted area information, reference policy information, reference season information, etc.) may also be applied.

[0165] Similarly, in these cases, when information corresponding to the reference information used as learning data is input, a solution search is performed based on the above-described method.

[0166] The search solution determined through the association may further be modified or weighted based on other reference information.

[0167] The other reference information referred to here corresponds to any reference information other than the reference information that is the base reference information, when any of the above-mentioned reference information is used as the base reference information.

[0168] For example, suppose that one of the other reference information, reference soil information P14, shows that futures prices were previously low. When new soil information corresponding to this reference soil information P14 is acquired, the system is set up in advance to perform a process to lower the weighting of search solutions with high futures prices, in other words, to lead to search solutions with low futures prices.

[0169] For example, assume that other reference information G is an analysis result suggesting a higher futures price level, and reference information F is an analysis result suggesting a lower futures price level. After setting the reference information in this way, if the actually acquired information is identical or similar to reference information G, a process is performed to increase the weighting of the higher futures price level. On the other hand, if the actually acquired information is identical or similar to reference information F, a process is performed to increase the weighting of the lower futures price level. In other words, the correlation itself leading to the futures price may be controlled based on this reference information F through H. Alternatively, the futures price may be determined based only on the above-mentioned correlation, and then the obtained search solution may be corrected based on the reference information F through H. In the latter case, the weighting and the degree of correction to the futures price as the search solution based on the reference information F through H will be reflected in the system design each time.

[0170] Furthermore, the reference information is not limited to being composed of any one type, and a solution search may be performed based on two or more types of reference information. Similarly, in such a case, the more the reference information suggests a higher futures price, the higher the futures price as the search solution obtained via the correlation may be corrected, and the more the reference information suggests a lower futures price, the lower the futures price as the search solution obtained via the correlation may be corrected.

[0171] As shown in Figure 19, the reference information that forms the basis of the correlation with futures increase / decrease data can be any reference information in the first or second embodiment (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, etc.).

[0172] For example, if reference natural environment information is used as the basic reference information, a correlation may be established between this and futures increase / decrease data.

[0173] Similarly, as shown in Figure 20, when forming a correlation between futures increase / decrease data and a combination of underlying reference information and other reference information, any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, etc.) can be applied as the underlying reference information. The other reference information includes any of the reference information in the first and second embodiments other than the underlying reference information.

[0174] In this case, if the basic reference information is the reference natural environment information, the other reference information includes any other reference information in the first embodiment and the second embodiment.

[0175] In such cases, by performing a similar solution search, it is possible to estimate the increase or decrease in the futures price.

[0176] In the second embodiment as well, the degree of association may be learned by combining not only one piece of other reference information but also two or more pieces of other reference information.

[0177] As the search solution described above, the quality of the crop or the yield of the crop may be searched for as a search solution instead of an increase or decrease in the price of the futures. Then, futures increase or decrease data corresponding to the crop quality or the yield of the crop as the search solution may be output. In such a case, a template is prepared in which the crop quality or the yield of the crop and the futures increase or decrease data are linked to each other, and by referencing this template, the futures increase or decrease data corresponding to the output solution of the crop quality or the yield of the crop is output.

[0178] In this case, data relating the quality of agricultural crops or the yield of agricultural crops to futures increase / decrease data through three or more levels of correlation may be prepared, so that futures increase / decrease data can be output in the same manner.

[0179] In the second embodiment, futures fluctuation data may be learned from the reference demand information as shown in Fig. 21. In such a case, three or more levels of correlation between the reference demand information and market price data of agricultural crop futures for which the agricultural crops grown in the past were harvested are used.

[0180] Reference demand information is data on the demand for agricultural products at the time of past transactions. Reference demand information includes any information that indicates the level of demand for agricultural products. Reference demand information may be composed of information on sales and inventory quantities of agricultural products at national or regional levels, and may be obtained from sales and inventory quantities at retail stores or wholesale markets. Reference demand information may also include data on the domestic or international production volume of the agricultural product. The greater the domestic or international production volume of the agricultural product, the lower the demand for the agricultural product; conversely, if production volume is low, the demand for the agricultural product increases. Reference demand information may also include data on the number of livestock that use the agricultural product as feed. This is because the demand for agricultural products as data for livestock increases when there is a large amount of livestock. Reference demand information may also include population estimate data. This is because the demand for agricultural products increases as the population increases.

[0181] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new market prices for agricultural products. In such cases, new demand information will be acquired.

[0182] The newly acquired demand information is information about the demand at the time of trading of agricultural products when displaying information about the market price of the agricultural product futures being traded. The content of this demand information is the same as that of the reference demand information. The specific method for estimating the market price of agricultural products is the same as that in the first embodiment, so a description thereof will be omitted below.

[0183] The market price changes depending on the actual demand for agricultural products at the time of trading. Therefore, by training a data set that shows the relationship between such reference demand information and the actual market price of agricultural products, it becomes possible to estimate the market price according to the demand for agricultural products.

[0184] In the second embodiment, futures fluctuation data may be learned from the reference supply information as shown in Fig. 22. In such a case, three or more levels of correlation between the reference supply information and market price data of agricultural crop futures for which the agricultural crops grown in the past were harvested are used.

[0185] Reference supply information is data on the supply status of agricultural products traded in the past. Reference supply information includes any information indicating the amount of agricultural product available. Reference supply information may be composed of information on the actual production volume of agricultural products at the national or regional level, or may be obtained from data on the amount of agricultural products delivered to retail stores or wholesale markets from production areas. Reference supply information may also be composed of one or more of the following: data on the planted area of ​​previously traded agricultural products; data on the farming history of the agricultural products; image data of the agricultural products taken during cultivation; survey data on the soil composition of the agricultural products; and natural environment data detecting weather and disaster conditions during the cultivation process of the agricultural products. The production volume of agricultural products changes depending on the planted area, farming history, and growth conditions that can be read from images of the agricultural products, soil composition, weather, and disaster conditions, which in turn changes the supply volume of agricultural products available on the market. As the supply volume of agricultural products available on the market changes, agricultural product futures also increase or decrease accordingly. For this reason, information on this supply volume is used as an explanatory variable.

[0186] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, it will be used to estimate new market prices for agricultural products. In such cases, new supply information will be acquired.

[0187] The newly acquired supply information is information about the supply at the time of trading of agricultural products when displaying information about the market price of the agricultural product futures being traded. The content of this supply information is the same as that of the reference supply information. The specific method for estimating the market price of agricultural products is the same as that in the first embodiment, so a description thereof will be omitted below.

[0188] The market price changes depending on the actual supply of agricultural products at the time of trading. Therefore, by training a data set showing the relationship between such reference supply information and the actual market price of agricultural products, it becomes possible to estimate the market price according to the supply of agricultural products.

[0189] The example in Figure 23 is premised on the formation of a combination of reference demand information and reference supply information. In the example in Figure 23, for example, reference demand information P11 to P13 and reference supply information P14 to P17 are assumed as input data. The intermediate nodes shown in Figure 23 are formed by combining the reference supply information with the reference demand information as input data. Each intermediate node is further connected to an output. In this output, the increase / decrease data of agricultural product futures is displayed as an output solution.

[0190] Each combination (intermediate node) of reference demand information and reference supply information is correlated with the agricultural product futures fluctuation data or market price as the output solution through three or more levels of correlation. The reference demand information and reference supply information are arranged on the left side via this correlation, and the agricultural product futures fluctuation data or market price are arranged on the right side via this correlation. The correlation indicates the degree to which the agricultural product futures fluctuation data or market price is highly related to the reference demand information and reference supply information arranged on the left side. In other words, this correlation is an index indicating the likelihood that each reference demand information and reference supply information is linked to which agricultural product futures fluctuation data or market price, and indicates the accuracy of selecting the most likely agricultural product futures fluctuation data or market price from the reference demand information and reference supply information. In the example of Figure 23, correlations w13 to w22 are shown.

[0191] The estimation device 2 acquires in advance three or more stages of correlation w13 to w22 as shown in Fig. 23. That is, when determining an actual search solution, the estimation device 2 accumulates reference demand information, reference supply information, and past data on the fluctuation data or market price of agricultural produce futures at that time, and creates the correlation shown in Fig. 23 by analyzing these.

[0192] For example, assume that the reference demand information is P11. In this case, if the supply situation at the time of obtaining the reference demand information is investigated and the reference supply information is P14, the amount of futures increase / decrease data or market price in the previous data is extracted.

[0193] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, when actually determining new increase / decrease data or market prices (market price data) for agricultural product futures, the trained data described above will be used to determine the increase / decrease data or market price for agricultural product futures. In such cases, new demand information and supply information will be acquired.

[0194] Based on the newly acquired demand information and supply information in this way, the fluctuation data or market price of the futures of the agricultural product for which the newly acquired demand information and supply information has actually been obtained is predicted. In such a case, the correlation degrees shown in FIG. 23 (Table 1) acquired in advance are referenced, and the one with the highest correlation degree is selected as the optimal solution. The details of the method for selecting this optimal solution are the same as those described in the first embodiment above.

[0195] Furthermore, when the reference demand information or reference supply information is used as the basic reference information shown in Figure 20, it may be combined with any of the information shown in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, etc.) as other reference information, and a correlation may be established between this and the futures increase / decrease data.

[0196] Furthermore, as shown in FIG. 18 , three or more levels of correlation between the underlying reference information and futures prices (increase / decrease data for each futures) may be used. In such a case, a solution search is performed based on three or more levels of correlation between the reference information according to newly acquired information and futures prices. The underlying reference information may also include reference supply information or reference demand information. Other reference information may include any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference agricultural work history information, reference expert opinion information, reference natural environment information, reference soil information, reference planted area information, reference policy information, reference season information, etc.), as well as reference supply information or reference demand information.

[0197] Third embodiment The third embodiment is a futures trading information display program that displays information about the market price of crude oil (crude oil or petroleum, gasoline, kerosene).

[0198] In the third embodiment, futures increase / decrease data may be learned from the reference demand information as shown in Fig. 21. In such a case, three or more levels of correlation between the reference demand information and the price data of crude oil futures traded in the past are used.

[0199] Reference demand information is data on the demand for crude oil at the time of past transactions. Reference demand information includes any information indicating the level of demand for crude oil. Reference demand information may be composed of information on crude oil sales volumes and stockpiles at national or regional levels, and may be obtained from sales volumes and stockpiles at gas stations, etc., and stockpiles of crude oil, petroleum, and gasoline stored in tanks. Reference demand information may also include data on the domestic or international production volume of the crude oil. The greater the domestic or international production volume of crude oil, the lower the demand for crude oil. Conversely, when production volume is low, the demand for crude oil increases. In addition, this reference demand information may be obtained from the situation of each industry that uses crude oil. For example, the greater the number of gasoline-using automobiles produced, sold, or on the road, the greater the demand for gasoline. It may also be obtained from the production volume and consumption trends of other petroleum-based petroleum products (synthetic resins, asphalt, fuel oil, paraffin, lubricating oil, etc.). We may also include various national policies toward crude oil in oil-producing and oil-consuming countries, as these affect oil demand.

[0200] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new crude oil trading prices. In such cases, new demand information will be acquired.

[0201] The newly acquired demand information is information about the demand at the time of crude oil trading when displaying information about the market price of the crude oil futures being traded. The content of this demand information is the same as that of the reference demand information. The specific method for estimating the market price of crude oil is the same as that of the first embodiment, so the explanation will be omitted below.

[0202] The trading price changes depending on the actual demand for crude oil at the time of trading. Therefore, by training a data set showing the relationship between such reference demand information and the actual trading price of crude oil, it becomes possible to estimate the trading price according to the demand for crude oil.

[0203] In the third embodiment, futures increase / decrease data may be learned from the reference supply information as shown in Fig. 22. In such a case, three or more levels of correlation between the reference supply information and the price data of crude oil futures traded in the past are used.

[0204] Reference supply information is data on the supply status of crude oil at the time of past transactions. Reference supply information includes any information indicating the amount of crude oil supply. Reference supply information may be composed of information on actual crude oil production volumes by country or region, or may be obtained from data on the amount of crude oil delivered to retail stores and gas stations from producing areas. Reference supply information may also include data on past crude oil production volumes, data on the production capacity of the crude oil, information on crude oil production technology and refining technology for the crude oil, and various national crude oil policies in oil-producing and crude oil-consuming countries, as these policies affect the oil supply situation.

[0205] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new crude oil trading prices. In such cases, new supply information will be acquired.

[0206] The newly acquired supply information is information about the supply at the time of crude oil trading when displaying information about the market price of the crude oil futures being traded. The content of this supply information is the same as that of the reference supply information. The specific method for estimating the market price of crude oil is the same as that of the first embodiment, so a description thereof will be omitted below.

[0207] The trading price changes depending on the actual supply of crude oil at the time of crude oil trading. Therefore, by training a data set showing the relationship between such reference supply information and the actual trading price of crude oil, it becomes possible to estimate the trading price according to the supply of crude oil.

[0208] Similarly, for oil futures, as shown in Figure 23, it is assumed that a combination of reference demand information and reference supply information is formed. In the example of Figure 23, for example, reference demand information P11 to P13 and reference supply information P14 to P17 are used as input data. The intermediate nodes shown in Figure 23 are formed by combining the reference demand information as input data with the reference supply information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for crude oil futures is displayed as an output solution.

[0209] Each combination (intermediate node) of reference demand information and reference supply information is correlated with the crude oil futures fluctuation data or market price as the output solution through three or more levels of correlation. The reference demand information and reference supply information are arranged on the left side via this correlation, and each crude oil futures fluctuation data or market price is arranged on the right side via the correlation. The correlation indicates the degree to which the reference demand information and reference supply information arranged on the left side are highly correlated with which crude oil futures fluctuation data or market price. In other words, this correlation is an index indicating the likelihood that each reference demand information and reference supply information is linked to which crude oil futures fluctuation data or market price, and indicates the accuracy of selecting the most likely crude oil futures fluctuation data or market price from the reference demand information and reference supply information. In the example of Figure 23, correlations w13 to w22 are shown.

[0210] The estimation device 2 acquires in advance three or more stages of correlation w13 to w22 as shown in Fig. 23. That is, when determining an actual search solution, the estimation device 2 accumulates reference demand information, reference supply information, and past data on the fluctuation data or market price of crude oil futures at that time, and creates the correlation shown in Fig. 23 by analyzing these.

[0211] For example, assume that the reference demand information is P11. In this case, if the supply situation at the time of obtaining the reference demand information is investigated and the reference supply information is P14, the amount of futures increase / decrease data or market price in the previous data is extracted.

[0212] This correlation is what is called "learned data" in artificial intelligence. After creating this learned data, when actually determining new data on increases / decreases or prices (price data) for crude oil futures, the learned data will be used to determine the data on increases / decreases or prices for crude oil futures. In such cases, new demand information and supply information will be acquired.

[0213] Based on the newly acquired demand information and supply information in this way, the fluctuation data or market price of the crude oil futures for which the newly acquired demand information and supply information is actually obtained is predicted. In such cases, the correlation degrees shown in FIG. 23 (Table 1) acquired in advance are referenced, and the one with the highest correlation degree is selected as the optimal solution. The details of the method for selecting this optimal solution are the same as the method described in the first embodiment above.

[0214] Furthermore, when the reference demand information or reference supply information is used as the basic reference information shown in Figure 20, it may be combined with any of the information shown in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, etc.) as other reference information, and a correlation may be established between this and the futures increase / decrease data.

[0215] Of these, the reference market information is added as a parameter because the market conditions themselves can affect actual crude oil prices, while the reference natural environment information is added as a parameter because a warm winter can reduce fuel demand.

[0216] Since reference external environmental information may also affect crude oil prices, it is added as a parameter.

[0217] The reference season information is also added as a parameter because the supply and demand balance differs depending on the season, such as the end of the month or the end of the year, when vehicle traffic volume is high, or the season, such as winter, when fuel demand is high.

[0218] Furthermore, the third embodiment is not limited to the above-described embodiments. For example, as shown in FIG. 18, three or more levels of correlation between the reference information serving as the base and futures prices may be used. In such a case, a solution search is performed based on three or more levels of correlation between the reference information according to newly acquired information and futures prices. The base reference information may be, for example, reference demand information or reference supply information. As other reference information, any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference agricultural work history information, reference expert opinion information, reference natural environment information, reference soil information, reference planted area information, reference policy information, reference seasonal information, etc.) may be applied. Furthermore, when the base reference information is reference demand information, other reference information may be reference supply information, or when the base reference information is reference supply information, other reference information may be reference demand information.

[0219] Similarly, in these cases, when information corresponding to the reference information used as learning data is input, a solution search is performed based on the above-described method.

[0220] The search solution determined through the association may further be modified or weighted based on other reference information.

[0221] The other reference information referred to here corresponds to any reference information other than the reference information that is the base reference information, when any of the above-mentioned reference information is used as the base reference information.

[0222] For example, assume that other reference information G is an analysis result suggesting a higher futures price level, and reference information F is an analysis result suggesting a lower futures price level. After setting the reference information in this way, if the actually acquired information is identical or similar to reference information G, a process is performed to increase the weighting of the higher futures price level. On the other hand, if the actually acquired information is identical or similar to reference information F, a process is performed to increase the weighting of the lower futures price level. In other words, the correlation itself leading to the futures price may be controlled based on this reference information F through H. Alternatively, the futures price may be determined based only on the above-mentioned correlation, and then the obtained search solution may be corrected based on the reference information F through H. In the latter case, the weighting and the degree of correction to the futures price as the search solution based on the reference information F through H will be reflected in the system design each time.

[0223] Furthermore, the reference information is not limited to being composed of any one type, and a solution search may be performed based on two or more types of reference information. Similarly, in such a case, the more the reference information suggests a higher futures price, the higher the futures price as the search solution obtained via the correlation may be corrected, and the more the reference information suggests a lower futures price, the lower the futures price as the search solution obtained via the correlation may be corrected.

[0224] Also, in the third embodiment, it goes without saying that the other reference information in FIGS. 18 and 20 may be combined with two or more other information to learn the degree of association.

[0225] Fourth embodiment The fourth embodiment is a futures trading information display program that displays information about rubber market prices.

[0226] In the fourth embodiment, futures increase / decrease data may be learned from the reference demand information as shown in Fig. 21. In such a case, three or more levels of correlation between the reference demand information and the price data of rubber futures traded in the past are used.

[0227] Reference demand information is data on the demand for rubber at the time of past transactions. Reference demand information includes any information indicating the level of demand for rubber. Reference demand information may be composed of information on the sales volume and inventory volume of rubber at the national or regional level, and may be obtained from sales and inventory volumes at retailers, wholesalers, etc. Reference demand information may also include data on the domestic or international production volume of the rubber. The greater the domestic or international production volume of rubber, the lower the demand for rubber. Conversely, the lower the production volume, the higher the demand for rubber. In addition, this reference demand information may be obtained from the situation of each industry that uses rubber. For example, the greater the production volume of automobiles or tires that use rubber, the higher the demand for rubber. Reference demand information may also be obtained from the production volume and consumption trends of other rubber products that use rubber (synthetic resins, rubber bands, etc.). Information on demand trends and domestic situations in rubber-consuming countries, as well as various domestic policies in rubber-consuming countries, may also be included, as these affect the demand for rubber.

[0228] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used when actually estimating new trading prices for rubber. In such cases, new demand information will be acquired.

[0229] The newly acquired demand information is information about the demand at the time of rubber trading when displaying information about the market price of the rubber futures being traded. The content of this demand information is the same as that of the reference demand information. The specific method for estimating the market price of rubber is the same as that in the first embodiment, so a description thereof will be omitted below.

[0230] The trading price changes depending on the actual demand for rubber at the time of trading. Therefore, by training a data set showing the relationship between such reference demand information and the actual trading price of rubber, it becomes possible to estimate the trading price according to the demand for rubber.

[0231] In the fourth embodiment, futures increase / decrease data may be learned from the reference supply information as shown in Fig. 22. In such a case, three or more levels of correlation between the reference supply information and the market price data of rubber futures traded in the past are used.

[0232] Reference supply information is data on the supply status of rubber at the time of past rubber transactions. Reference supply information includes any information that indicates the amount of rubber supply. Reference supply information may be composed of information on the actual production volume of rubber at the national or regional level, or may be obtained from data on the amount of rubber delivered to retailers and wholesalers from producing areas. Reference supply information may also include data on past rubber production volumes, data on the production capacity of the rubber, information on rubber production technology for the rubber, information on natural rubber production volumes, weather in rubber-producing countries, domestic conditions in rubber-producing countries, and various policies regarding rubber in rubber-producing countries, as these factors affect the rubber supply status.

[0233] This correlation is what artificial intelligence calls learned data. After creating this learned data, the learned data will be used when actually estimating new rubber market prices. In such cases, new supply information will be acquired.

[0234] The newly acquired supply information is information about the supply at the time of rubber trading when displaying information about the market price of the rubber futures being traded. The content of this supply information is the same as that of the reference supply information. The specific method for estimating the market price of rubber is the same as that of the first embodiment, so a detailed explanation will be omitted below.

[0235] The trading price changes depending on the actual supply of rubber at the time of trading. Therefore, by training a data set showing the relationship between such reference supply information and the actual trading price of rubber, it becomes possible to estimate the trading price depending on the supply of rubber.

[0236] Similarly, as shown in Figure 23, it is assumed that a combination of reference demand information and reference supply information is formed for rubber futures. In the example of Figure 23, the input data is, for example, reference demand information P11 to P13 and reference supply information P14 to P17. The intermediate nodes shown in Figure 23 are formed by combining the reference demand information as input data with the reference supply information. Each intermediate node is further connected to an output. In this output, the increase / decrease data for rubber futures is displayed as an output solution.

[0237] Each combination (intermediate node) of reference demand information and reference supply information is correlated with the rubber futures fluctuation data or market price as the output solution through three or more levels of correlation. The reference demand information and reference supply information are arranged on the left side via this correlation, and each rubber futures fluctuation data or market price is arranged on the right side via the correlation. The correlation indicates the degree to which rubber futures fluctuation data or market price is highly correlated with the reference demand information and reference supply information arranged on the left side. In other words, this correlation is an index indicating the likelihood that each reference demand information and reference supply information is linked to which rubber futures fluctuation data or market price, and indicates the accuracy of selecting the most likely rubber futures fluctuation data or market price from the reference demand information and reference supply information. In the example of Figure 23, correlations w13 to w22 are shown.

[0238] The estimation device 2 acquires in advance three or more stages of correlation w13 to w22 as shown in Fig. 23. That is, when determining an actual search solution, the estimation device 2 accumulates reference demand information, reference supply information, and past data on the increase / decrease data or market price of rubber futures at that time, and creates the correlation shown in Fig. 23 by analyzing these.

[0239] For example, assume that the reference demand information is P11. In this case, if the supply situation at the time of obtaining the reference demand information is investigated and the reference supply information is P14, the amount of futures increase / decrease data or market price in the previous data is extracted.

[0240] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, when actually determining new increase / decrease data or market prices (market price data) for rubber futures, the trained data described above will be used to determine the increase / decrease data or market price for rubber futures. In such cases, new demand information and supply information will be acquired.

[0241] Based on the newly acquired demand information and supply information in this way, the fluctuation data or market price of the rubber futures for which the newly acquired demand information and supply information are actually obtained is predicted. In such cases, the correlation degrees shown in Figure 23 (Table 1) acquired in advance are referenced, and the one with the highest correlation degree is selected as the optimal solution. The details of the method for selecting this optimal solution are the same as the method described in the first embodiment above.

[0242] Furthermore, when the reference demand information or the reference supply information is used as the basic reference information shown in FIG. 20 , any of the information shown in the first to third embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, etc.) may be combined as other reference information to form a correlation between this and futures increase / decrease data. In addition to these, reference crude oil price information related to the crude oil price at the time of past rubber transactions may also be combined as other reference information. Because crude oil is the raw material for rubber, the market price and supply / demand balance of rubber change depending on the crude oil price. Therefore, this reference crude oil price information may also be added to the explanatory variables.

[0243] Of these, the reference market information is added as a parameter because the market conditions themselves may affect the actual rubber price.

[0244] Since reference external environmental information may also affect rubber prices, it is added as a parameter.

[0245] The reference season information is also added as a parameter because the supply and demand balance varies depending on the season, for example, at the end of the month or the end of the year, when vehicle traffic volume is high, or when demand for rubber increases due to certain events.

[0246] Furthermore, the fourth embodiment is not limited to the above-described embodiments. For example, as shown in FIG. 18, three or more levels of correlation between the reference information serving as the base and futures prices may be used. In such a case, a solution search is performed based on three or more levels of correlation between the reference information according to newly acquired information and futures prices. The base reference information may be, for example, reference demand information or reference supply information. As other reference information, any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference agricultural work history information, reference expert opinion information, reference natural environment information, reference soil information, reference planted area information, reference policy information, reference season information, reference crude oil price information, etc.) may be applied. Furthermore, when the base reference information is reference demand information, other reference information may be reference supply information, or when the base reference information is reference supply information, other reference information may be reference demand information.

[0247] Similarly, in these cases, when information corresponding to the reference information used as learning data is input, a solution search is performed based on the above-described method.

[0248] The search solution determined through the association may further be modified or weighted based on other reference information.

[0249] The other reference information referred to here corresponds to any reference information other than the reference information that is the base reference information, when any of the above-mentioned reference information is used as the base reference information.

[0250] For example, assume that other reference information G is an analysis result suggesting a higher futures price level, and reference information F is an analysis result suggesting a lower futures price level. After setting the reference information in this way, if the actually acquired information is identical or similar to reference information G, a process is performed to increase the weighting of the higher futures price level. On the other hand, if the actually acquired information is identical or similar to reference information F, a process is performed to increase the weighting of the lower futures price level. In other words, the correlation itself leading to the futures price may be controlled based on this reference information F through H. Alternatively, the futures price may be determined based only on the above-mentioned correlation, and then the obtained search solution may be corrected based on the reference information F through H. In the latter case, the weighting and the degree of correction to the futures price as the search solution based on the reference information F through H will be reflected in the system design each time.

[0251] Furthermore, the reference information is not limited to being composed of any one type, and a solution search may be performed based on two or more types of reference information. Similarly, in such a case, the more the reference information suggests a higher futures price, the higher the futures price as the search solution obtained via the correlation may be corrected, and the more the reference information suggests a lower futures price, the lower the futures price as the search solution obtained via the correlation may be corrected.

[0252] Also, in the fourth embodiment, it goes without saying that the other reference information in FIGS. 18 and 20 may be combined with two or more other information to learn the degree of association.

[0253] Fifth embodiment The fifth embodiment is a futures trading information display program that displays information about the market prices of precious metals.

[0254] In the fifth embodiment, futures increase / decrease data may be learned from the reference demand information as shown in Fig. 21. In such a case, three or more levels of correlation between the reference demand information and the price data of precious metal futures traded in the past are used.

[0255] Reference demand information is data on the demand for precious metals at the time of past transactions. Precious metals include all precious metals that are subject to futures trading, such as gold, silver, platinum, palladium, and rare metals. Reference demand information includes any information indicating the level of demand for precious metals. Reference demand information may be composed of information on sales and inventory of precious metals at national or regional levels, and may be obtained from sales and inventory at retailers, wholesalers, etc. Reference demand information may also include data on domestic or international production of the precious metal. The greater the domestic or international production of a precious metal, the lower the demand for that metal. Conversely, when production is low, the demand for that metal increases. In addition, this reference demand information may be obtained from the situation of each industry that uses precious metals. For example, the greater the production volume of devices and electronic devices that use precious metals, the higher the demand for precious metals. Reference demand information may also be obtained from the production volume and consumption trends of precious metal products (such as jewelry) that use precious metals. Precious metals may also be used in production processes. Platinum is sometimes used as a catalyst, and the addition of this catalyst is required in the production process. Demand for jewelry, chemical products, glass products, etc. may also be included in this reference demand information. The reference demand information may also include information on demand trends and domestic situations in consuming countries, as well as various domestic policies in consuming countries of precious metals, as these also affect the demand for precious metals.

[0256] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new precious metals trading prices. In such cases, new demand information will be acquired.

[0257] The newly acquired demand information is information about the demand at the time of precious metal trading when displaying information about the market price of the precious metal futures being traded. The content of this demand information is the same as the reference demand information. The specific method for estimating the trading market price of precious metals is the same as in the first embodiment, so the explanation below will be omitted.

[0258] The trading price of precious metals changes depending on the actual demand for the metals when they are traded. Therefore, by training a data set that shows the relationship between such reference demand information and the actual trading price of precious metals, it becomes possible to estimate the trading price according to the demand for precious metals.

[0259] In the fifth embodiment, futures fluctuation data may be learned from the reference supply information as shown in Fig. 22. In such a case, three or more levels of correlation between the reference supply information and the price data of precious metal futures traded in the past are used.

[0260] Reference supply information is data on the supply status of precious metals at the time of past transactions. Reference supply information includes any information indicating the supply level of precious metals. Reference supply information may be composed of information on the actual production volume of precious metals at the national or regional level, or may be obtained from data on the amount of precious metals delivered to retailers and wholesalers from production areas. Reference supply information may also include data on past production volumes of precious metals, data on the production capacity and refining capacity of the precious metals, information on the precious metal production technology for the precious metals, information on the production volume of natural precious metal ore, weather in precious metal-producing countries, domestic conditions in precious metal-producing countries, and various policies regarding precious metals in precious metal-producing countries, as these factors affect the supply status of precious metals. Reference supply information may also include information on the production costs of precious metals, information on mines that produce precious metals, etc.

[0261] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new precious metals trading prices. In such cases, new supply information will be acquired.

[0262] The newly acquired supply information is information about the supply at the time of precious metal trading when displaying information about the market price of precious metal futures being traded. The content of this supply information is the same as the reference supply information. The specific method for estimating the market price of precious metals is the same as in the first embodiment, so a detailed explanation will be omitted below.

[0263] The trading price of precious metals changes depending on the actual supply of precious metals when they are traded. Therefore, by training a data set that shows the relationship between such reference supply information and the actual trading price of precious metals, it becomes possible to estimate the trading price according to the supply of precious metals.

[0264] Similarly, for precious metal futures, as shown in Figure 23, it is assumed that a combination of reference demand information and reference supply information is formed. In the example of Figure 23, the input data is, for example, reference demand information P11 to P13 and reference supply information P14 to P17. The intermediate nodes shown in Figure 23 are formed by combining the reference supply information with the reference demand information as input data. Each intermediate node is further connected to an output. In this output, the increase / decrease data of precious metal futures is displayed as an output solution.

[0265] Each combination (intermediate node) of reference demand information and reference supply information is correlated with the precious metal futures fluctuation data or market price as the output solution through three or more levels of correlation. The reference demand information and reference supply information are arranged on the left side via this correlation, and each precious metal futures fluctuation data or market price is arranged on the right side via the correlation. The correlation indicates the degree to which precious metal futures fluctuation data or market price is highly correlated with the reference demand information and reference supply information arranged on the left side. In other words, this correlation is an index indicating which precious metal futures fluctuation data or market price each reference demand information and reference supply information is likely to be linked to, and indicates the accuracy of selecting the most likely precious metal futures fluctuation data or market price from the reference demand information and reference supply information. In the example of Figure 23, correlations w13 to w22 are shown.

[0266] The estimation device 2 acquires in advance three or more stages of correlation w13 to w22 as shown in Fig. 23. That is, when determining an actual search solution, the estimation device 2 accumulates reference demand information, reference supply information, and past data on the fluctuation data or market price of precious metal futures at that time, and creates the correlation shown in Fig. 23 by analyzing these.

[0267] For example, assume that the reference demand information is P11. In this case, if the supply situation at the time of obtaining the reference demand information is investigated and the reference supply information is P14, the amount of futures increase / decrease data or market price in the previous data is extracted.

[0268] This correlation is what is called "learned data" in artificial intelligence. After creating this learned data, when actually determining new fluctuation data or market prices (market price data) for precious metal futures, the learned data described above will be used to determine the fluctuation data or market price for precious metal futures. In such cases, new demand information and supply information will be acquired.

[0269] Based on the newly acquired demand information and supply information in this way, the fluctuation data or market price of the precious metal futures for which the newly acquired demand information and supply information is actually obtained is predicted. In such cases, the correlation degrees shown in Figure 23 (Table 1) acquired in advance are referenced, and the one with the highest correlation degree is selected as the optimal solution. The details of the method for selecting this optimal solution are the same as the method described in the first embodiment above.

[0270] Furthermore, when the reference demand information or reference supply information is used as the basic reference information shown in Figure 20, it may be combined with any of the information shown in the first to fourth embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, etc.) as other reference information, and a correlation may be established between this and the futures increase / decrease data.

[0271] Other reference information may also be combined with reference price information relating to prices at the time of precious metal transactions in the past. Prices have a significant impact on precious metal market prices, and the market prices and supply-demand balance of precious metals change depending on prices, so this reference price information may also be added to the explanatory variables. In such cases, price information relating to prices at the time when a new increase or decrease in the price of precious metals is to be estimated is obtained.

[0272] Other reference information may include reference public institution holdings information regarding the amount of precious metals held by public institutions at the time of past precious metals transactions. The amount of precious metals held by public institutions has a significant impact on market prices, so this may be included as an explanatory variable. In such cases, public institution holdings information regarding the amount of precious metals held by public institutions is obtained when a new increase or decrease in the price of precious metals is to be estimated.

[0273] Other reference information may include reference international situation information relating to the international situation at the time of past precious metal transactions. Precious metals are often imported and exported between countries, and international situations have a significant impact on market prices, so they may be included as explanatory variables. In such cases, international situation information relating to the international situation at the time when new increases or decreases in precious metal prices are to be estimated is obtained.

[0274] Of these, the reference market information is added as a parameter because the market conditions themselves can sometimes affect actual precious metal prices.

[0275] Since reference external environmental information may also affect precious metal prices, it is added as a parameter.

[0276] Furthermore, the fifth embodiment is not limited to the above-described embodiments. For example, as shown in FIG. 18, three or more levels of correlation between the reference information serving as the base and futures prices may be used. In such a case, a solution search is performed based on three or more levels of correlation between the reference information according to newly acquired information and futures prices. The base reference information is, for example, reference demand information or reference supply information. Other reference information may include any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference agricultural work history information, reference expert opinion information, reference natural environment information, reference soil information, reference planted area information, reference policy information, reference season information, reference crude oil price information, reference price information, reference public institution holdings information, reference international situation information, etc.). In addition, when the basic reference information is reference demand information, other reference information may be reference supply information, and when the basic reference information is reference supply information, other reference information may be reference demand information.

[0277] Similarly, in these cases, when information corresponding to the reference information used as learning data is input, a solution search is performed based on the above-described method.

[0278] The search solution determined through the association may further be modified or weighted based on other reference information.

[0279] The other reference information referred to here corresponds to any reference information other than the reference information that is the base reference information, when any of the above-mentioned reference information is used as the base reference information.

[0280] For example, assume that other reference information G is an analysis result suggesting a higher futures price level, and reference information F is an analysis result suggesting a lower futures price level. After setting the reference information in this way, if the actually acquired information is identical or similar to reference information G, a process is performed to increase the weighting of the higher futures price level. On the other hand, if the actually acquired information is identical or similar to reference information F, a process is performed to increase the weighting of the lower futures price level. In other words, the correlation itself leading to the futures price may be controlled based on this reference information F through H. Alternatively, the futures price may be determined based only on the above-mentioned correlation, and then the obtained search solution may be corrected based on the reference information F through H. In the latter case, the weighting and the degree of correction to the futures price as the search solution based on the reference information F through H will be reflected in the system design each time.

[0281] Furthermore, the reference information is not limited to being composed of any one type, and a solution search may be performed based on two or more types of reference information. Similarly, in such a case, the more the reference information suggests a higher futures price, the higher the futures price as the search solution obtained via the correlation may be corrected, and the more the reference information suggests a lower futures price, the lower the futures price as the search solution obtained via the correlation may be corrected.

[0282] Also, in the fifth embodiment, it goes without saying that the other reference information in FIGS. 18 and 20 may be combined with two or more other information to learn the degree of association.

[0283] Sixth embodiment The sixth embodiment is a futures trading information display program that displays information about market prices of marine products.

[0284] In the sixth embodiment, futures fluctuation data may be learned from the reference demand information as shown in Fig. 21. In such a case, three or more levels of correlation between the reference demand information and market price data of marine product futures traded in the past are used.

[0285] Reference demand information is data on the demand for a seafood product at the time of past transactions. Examples of seafood include fish, shellfish, seaweed, seaweed (dried squid, kelp, fish oil, fish meal, etc.), and all seafood products subject to futures trading. Reference demand information includes any information indicating the level of demand for a seafood product. Reference demand information may be composed of information on sales volume and inventory volume of seafood at a national or regional level, and may be obtained from sales numbers and inventory volumes at retail stores, fish markets, wholesalers, etc. Reference demand information may also include data on the domestic or international production volume of the seafood product. The greater the domestic or international production volume of a seafood product, the lower the demand for that seafood product. Conversely, when production volume is low, the demand for the seafood product increases. In addition, this reference demand information may be obtained from the situation of each industry that uses seafood. For example, the greater the production volume of products that use fish meal as a seafood product, the higher the demand for the seafood product. It may also be obtained from the production volume and consumption trends of other seafood products (such as cut squid) that use seafood. The reference demand information may also include information on demand trends and domestic situations in consuming countries, as well as various domestic policies in consuming countries of seafood products, as these also affect the demand for seafood.

[0286] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new market prices for seafood. In such cases, new demand information will be acquired.

[0287] The newly acquired demand information is information about the demand at the time of trading of the seafood when displaying information about the market price of the seafood futures being traded. The content of this demand information is the same as that of the reference demand information. The specific method for estimating the market price of the seafood is the same as that in the first embodiment, and therefore will not be described below.

[0288] The trading price of a seafood product changes depending on the actual demand for that product at the time of trading. Therefore, by training a data set that shows the relationship between such reference demand information and the actual trading price of the seafood product, it becomes possible to estimate the trading price according to the demand for the seafood product.

[0289] In the sixth embodiment, futures fluctuation data may be learned from the reference supply information as shown in Fig. 22. In such a case, three or more levels of correlation between the reference supply information and market price data of marine product futures traded in the past are used.

[0290] Reference supply information is data on the supply status of a seafood product at the time of past transactions. Reference supply information includes any information indicating the supply level of the seafood product. Reference supply information may be composed of information on the actual production volume of the seafood product at the national or regional level, or may be obtained from data on the amount of seafood delivered to retailers and wholesalers from production areas. Reference supply information may also include data on past production volumes of the seafood product, data on the processing capacity of the seafood product, information on the seafood processing technology for the seafood product, information on the seafood catch volume, weather in the seafood-producing country, domestic conditions in the seafood-producing country, and various policies regarding seafood in the seafood-producing country, as these factors affect the supply status of the seafood product. Furthermore, reference supply information may also include information on the processing costs of the seafood product, information on the fishing grounds and aquaculture facilities that produce the seafood product, and, in the case of aquaculture, the type and amount of feed used, as well as the rearing environment. Furthermore, this reference supply information also includes the weather in the sea area where the marine products are landed (supply sea area) and the political situation of the country that includes the supply sea area in its territorial waters.

[0291] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, the trained data will be used to estimate new market prices for seafood. In such cases, new supply information will be acquired.

[0292] The newly acquired supply information is information about the supply at the time of trading of the seafood when displaying information about the market price of the seafood futures being traded. The content of this supply information is the same as that of the reference supply information. The specific method for estimating the market price of the seafood is the same as that in the first embodiment, so a description thereof will be omitted below.

[0293] The market price changes depending on the actual supply of seafood at the time of trading. Therefore, by training a data set showing the relationship between such reference supply information and the actual market price of seafood, it becomes possible to estimate the market price according to the supply of seafood.

[0294] Similarly, for seafood futures, it is assumed that a combination of reference demand information and reference supply information is formed, as shown in Figure 23. In the example of Figure 23, the input data is, for example, reference demand information P11 to P13 and reference supply information P14 to P17. The intermediate nodes shown in Figure 23 are formed by combining the reference demand information as input data with the reference supply information. Each intermediate node is further linked to an output. In this output, the increase / decrease data for seafood futures is displayed as an output solution.

[0295] Each combination (intermediate node) of reference demand information and reference supply information is interconnected with the seafood futures fluctuation data or market price as the output solution through three or more levels of correlation. The reference demand information and reference supply information are arranged on the left side via this correlation, and the seafood futures fluctuation data or market price for each seafood product are arranged on the right side via this correlation. The correlation indicates the degree to which seafood futures fluctuation data or market price is highly related to the reference demand information and reference supply information arranged on the left side. In other words, this correlation is an index indicating the likelihood that each reference demand information and reference supply information will be linked to which seafood futures fluctuation data or market price, and indicates the accuracy of selecting the most likely seafood futures fluctuation data or market price from the reference demand information and reference supply information. In the example of Figure 23, correlations w13 to w22 are shown.

[0296] The estimation device 2 acquires in advance three or more stages of correlation w13 to w22 as shown in Fig. 23. That is, when determining an actual search solution, the estimation device 2 accumulates reference demand information, reference supply information, and past data on fluctuations or market prices of seafood futures at that time, and analyzes these to create the correlations shown in Fig. 23.

[0297] For example, assume that the reference demand information is P11. In this case, if the supply situation at the time of obtaining the reference demand information is investigated and the reference supply information is P14, the amount of futures increase / decrease data or market price in the previous data is extracted.

[0298] This correlation is what is called "trained data" in artificial intelligence. After creating this trained data, when actually determining new increase / decrease data or market prices (market price data) for seafood futures, the trained data described above will be used to determine the increase / decrease data or market price for seafood futures. In such cases, new demand information and supply information will be acquired.

[0299] Based on the newly acquired demand information and supply information in this way, the fluctuation data or market price of the futures of the seafood product for which the newly acquired demand information and supply information has actually been obtained is predicted. In such cases, the correlation degrees shown in Figure 23 (Table 1) acquired in advance are referenced, and the one with the highest correlation degree is selected as the optimal solution. The details of the method for selecting this optimal solution are the same as those described in the first embodiment above.

[0300] Furthermore, when the reference demand information or reference supply information is used as the basic reference information shown in Figure 20, it may be combined with any of the information shown in the first to fifth embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, reference natural environment information, reference soil information, reference agricultural work history information, reference planted area information, reference policy information, reference season information, reference price information, international situation information, etc.) as other reference information, and a correlation may be established between this and the futures increase / decrease data.

[0301] Of these, the reference market information is added as a parameter because the market conditions themselves can sometimes affect actual fishery product prices.

[0302] Since reference external environmental information may also affect seafood prices, it is added as a parameter.

[0303] Furthermore, the sixth embodiment is not limited to the above-described embodiments. For example, as shown in FIG. 18, three or more levels of correlation between the underlying reference information and futures prices may be used. In such a case, a solution search is performed based on three or more levels of correlation between the reference information according to newly acquired information and futures prices. The underlying reference information is, for example, reference demand information or reference supply information. Other reference information may include any of the reference information in the first and second embodiments (reference market information, reference event information, reference external environment information, reference household information, reference real estate information, reference agricultural work history information, reference expert opinion information, reference natural environment information, reference soil information, reference planted area information, reference policy information, reference season information, reference crude oil price information, reference price information, reference public institution holdings information, reference international situation information, etc.). In addition, when the basic reference information is reference demand information, other reference information may be reference supply information, and when the basic reference information is reference supply information, other reference information may be reference demand information.

[0304] Similarly, in these cases, when information corresponding to the reference information used as learning data is input, a solution search is performed based on the above-described method.

[0305] The search solution determined through the association may further be modified or weighted based on other reference information.

[0306] The other reference information referred to here corresponds to any reference information other than the reference information that is the base reference information, when any of the above-mentioned reference information is used as the base reference information.

[0307] For example, assume that other reference information G is an analysis result suggesting a higher futures price level, and reference information F is an analysis result suggesting a lower futures price level. After setting the reference information in this way, if the actually acquired information is identical or similar to reference information G, a process is performed to increase the weighting of the higher futures price level. On the other hand, if the actually acquired information is identical or similar to reference information F, a process is performed to increase the weighting of the lower futures price level. In other words, the correlation itself leading to the futures price may be controlled based on this reference information F through H. Alternatively, the futures price may be determined based only on the above-mentioned correlation, and then the obtained search solution may be corrected based on the reference information F through H. In the latter case, the weighting and the degree of correction to the futures price as the search solution based on the reference information F through H will be reflected in the system design each time.

[0308] Furthermore, the reference information is not limited to being composed of any one type, and a solution search may be performed based on two or more types of reference information. Similarly, in such a case, the more the reference information suggests a higher futures price, the higher the futures price as the search solution obtained via the correlation may be corrected, and the more the reference information suggests a lower futures price, the lower the futures price as the search solution obtained via the correlation may be corrected.

[0309] Also, in the sixth embodiment, it goes without saying that the other reference information in FIGS. 18 and 20 may be combined with two or more other information to learn the degree of association. [Explanation of symbols]

[0310] 1. Futures Trading Information Display System 2 Search device 21 Internal Bus 23 Display section 24 Control Unit 25 Control section 26 Communications Department 27 Estimation part 28 Memory section 61 nodes

Claims

1. In a futures trading information display program that displays information on the market prices of precious metal futures to be traded, an information acquisition step of acquiring demand information regarding demand at the time of trading of precious metals; and a display step of displaying the price data of precious metal futures traded in the past by using a correlation of three or more levels between reference demand information relating to the demand for the precious metal at the time of trading the precious metal and price data of precious metal futures traded in the past, and giving priority to the reference demand information corresponding to the demand information acquired in the information acquisition step and price data of precious metal futures with a higher correlation of three or more levels. A futures trading information display program characterized by:

2. In the information acquisition step, supply information regarding the supply status at the time of the precious metal transaction is further acquired, In the display step, a combination of the reference demand information and reference supply information relating to the supply status of the precious metal at the time of the precious metal transaction in the past is used to display the price data of the precious metal futures at a correlation level of three or more, and the price data of the precious metal futures is displayed by giving priority to a combination of the reference demand information corresponding to the demand information acquired in the information acquisition step and the reference supply information corresponding to the supply information, the combination having a higher correlation level of three or more with the price data of the precious metal futures.

2. The futures trading information display program according to claim 1,

3. In the information acquisition step, as the supply information, any one or more of information on the production cost of precious metals, the domestic situation of precious metal producing countries, the policies of precious metal producing countries, and the situation of mines that produce precious metals are acquired; In the display step, the reference supply information may be any one or more of information on the production costs of precious metals traded in the past, the domestic situation of a precious metal producing country, the policy of a precious metal producing country, and the policy of a mine producing precious metals.

3. The futures trading information display program according to claim 2,

4. In the information acquisition step, market information regarding the market conditions at the time of the new futures transaction is further acquired, In the display step, a combination of the reference demand information and reference market information relating to market conditions at the time of trading of the precious metals traded in the past is used to display the price data of the precious metal futures at a correlation level of three or more, and the price data of the precious metal futures is displayed by giving priority to a combination of the reference demand information corresponding to the demand information acquired in the information acquisition step and the reference market information corresponding to the market information, the combination having a higher correlation level of three or more with the price data of the precious metal futures.

3. The futures trading information display program according to claim 1 or 2,

5. In the information acquisition step, external environment information reflecting the external environment at the time of new futures trading is further acquired, In the display step, a combination of the reference demand information and reference external environment information reflecting the external environment at the time of trading of the precious metals traded in the past is used to display the price data of the precious metal futures at a correlation level of three or more, and the price data of the precious metal futures is displayed by giving priority to a combination of the reference demand information corresponding to the demand information acquired in the information acquisition step and the reference external environment information corresponding to the external environment information, the combination having a higher correlation level of three or more with the price data of the precious metal futures.

3. The futures trading information display program according to claim 1 or 2,

6. In the information acquisition step, price information regarding prices at the time of new futures transactions is further acquired, In the display step, a combination of the reference demand information and reference price information relating to prices at the time of precious metal transactions in the past is used to display the market price data of the precious metal futures at a correlation level of three or more, and a combination of the reference demand information according to the demand information acquired in the information acquisition step and the reference price information according to the prices is given priority over a combination having a higher correlation level of three or more with the market price data of the precious metal futures.

2. The futures trading information display program according to claim 1,

7. In the information acquisition step, public institution holdings information regarding the amount of precious metals held by the public institution at the time of new futures trading is further acquired; In the display step, a combination of the reference demand information and reference public institution holding information relating to the amount of precious metal held by a public institution at the time of past precious metal transactions is used to display the price data of the precious metal futures at a correlation level of three or more, and the price data of the precious metal futures is displayed by giving priority to a combination of the reference demand information corresponding to the demand information acquired in the information acquisition step and the reference public institution holding information corresponding to the public institution holding information, which has a higher correlation level of three or more with the price data of the precious metal futures.

2. The futures trading information display program according to claim 1,

8. In the information acquisition step, further acquire information on international situations regarding the international situation at the time of new futures transactions, In the display step, a combination of the reference demand information and reference international situation information relating to the international situation at the time of precious metal transactions in the past is used to display the price data of the precious metal futures at a correlation level of three or more, and the price data of the precious metal futures is displayed by giving priority to a combination of the reference demand information corresponding to the demand information acquired in the information acquisition step and the reference international situation information corresponding to the international situation information, the combination having a higher correlation level of three or more with the price data of the precious metal futures.

2. The futures trading information display program according to claim 1,

9. In a futures trading information display program that displays information on the market prices of precious metal futures to be traded, an information acquisition step of acquiring demand information regarding demand at the time of trading of precious metals and supply information regarding supply status at the time of trading of the precious metals; a display step of using a correlation degree of three or more levels between reference demand information regarding the demand for precious metals at the time of trading of the precious metals traded in the past and market price data of precious metal futures traded in the past, giving priority to reference demand information corresponding to the demand information acquired in the information acquisition step and market price data of precious metal futures with a higher correlation degree of three or more levels, and displaying the market price data of precious metal futures based on the acquired supply information, A futures trading information display program characterized by:

Images