Server and computer program

Abstract

We provide technology that keeps transaction rates within a certain range in response to fluctuations in exchange rates. [Solution] A server that provides information regarding currency hedging to a user terminal, which uses multiple AI models to obtain predictions for each AI model of future exchange rate trends, accepts specification of a first amount of a first currency, accepts input of at least one of a first desired rate and a second desired rate for an exchange rate reservation of the first amount, and input of the expiration date of the exchange rate reservation, calculates a first realization probability for the period up to the expiration date of a limit order calculated from the first desired rate based on the predicted future trend of the exchange rate, calculates a second realization probability for the period up to the expiration date of a stop-loss order calculated from the second desired rate based on the predicted future trend of the exchange rate, and transmits information on the second amount to be subject to currency hedging to the user terminal along with information on the limit order and the first realization probability, and information on the stop-loss order and the second realization probability.

Description

[Technical Field]

[0001] The present invention relates to a server and a computer program. [Background technology]

[0002] Generally, when the price of a product involves foreign exchange (for example, when payment is made in Japanese yen but is expected to be converted into dollars, or vice versa), if there are fluctuations in the exchange rate between the time the product's sales price is set and the time the transaction is completed, the amount actually paid may differ from the amount expected when converted into foreign currency.

[0003] To prevent this situation, it is possible to use a mechanism known as a forward exchange contract (see Patent Document 1). By using a forward exchange contract, even if a certain period of time has passed between the time the purchase and sale price was set and the time the transaction was concluded, the payment amount is converted into cash at the exchange rate at the time the price was set, so there is no impact from exchange rate fluctuations during that certain period. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2019-220108 Summary of the Invention [Problem to be solved by the invention]

[0005] When making a forward exchange contract, it is expected that future market movements will be predicted and that one will be prepared to deal with situations in which market conditions fluctuate significantly beyond past information.

[0006] Therefore, the present invention provides a technique that enables the transaction rate to be kept within a certain range in response to fluctuations in exchange rates. [Means for solving the problem]

[0007] According to one aspect of the present invention for solving the above problem, there is provided a server that provides information on currency hedging to a user terminal, the server comprising: one or more processors; Memory and a program stored in the memory that, when executed by the one or more processors, Using a plurality of AI models, obtaining, for each AI model, a forecast of future changes in an exchange rate indicating the value of a second currency relative to a first currency; accepting a designation of a first amount of the first currency from the user terminal; receiving input of at least one of a first desired exchange rate and a second desired exchange rate for the first amount of foreign exchange contract, and an input of a maturity date of the foreign exchange contract; calculating a limit price for the exchange rate for a forward foreign exchange contract of the first amount based on the received first desired rate, the first amount, and a hedged amount already hedged in relation to the maturity date; calculating a stop loss of the exchange rate for the forward exchange contract of the first amount based on the received second desired rate, the first amount, and the hedged amount; Calculating a first realization probability of the calculated limit price during the period up to the expiration date based on future trends in the exchange rate individually predicted by the plurality of AI models; Calculating a second realization probability of the calculated stop loss order during the period up to the expiration date based on future trends in the exchange rate individually predicted by the plurality of AI models; and transmitting information on a second amount of the first amount that should be subject to currency hedging to the user terminal together with information on the calculated limit order and the calculated first realization probability, and information on the calculated stop order and the calculated second realization probability. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide a technique that makes it possible to keep transaction rates within a certain range in response to fluctuations in exchange rates. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of a system 10 according to an embodiment. [Figure 2] FIG. 1 is a diagram showing an example of a hardware configuration of an information processing apparatus according to an embodiment. [Figure 3] 4 is a timing chart corresponding to an example of processing executed in the system 10 according to the embodiment. [Figure 4A] FIG. 4 is a diagram showing an example of a main screen according to the embodiment. [Figure 4B] FIG. 10 is a diagram showing an example of a setting screen in a rate defense guide screen according to the embodiment. [Figure 4C] FIG. 10 is a diagram illustrating an example of a foreign currency payment and receipt plan input screen on a rate defense guide screen according to an embodiment. [Figure 4D] FIG. 10 is a diagram showing an example of a rate defense guide setting input screen on a rate defense guide screen according to the embodiment. [Figure 4E] FIG. 10 is a diagram showing an example of a result display of a rate defense guide according to an embodiment. [Figure 4F] FIG. 10 is a diagram showing another example of a result display of a rate defense guide according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, the embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the invention as claimed, and not all combinations of features described in the embodiments are necessarily essential to the invention. Two or more of the features described in the embodiments may be arbitrarily combined. Furthermore, the same reference numerals are used for the same or similar components, and redundant description will be omitted.

[0011] <System configuration> FIG. 1 shows a general configuration of a system 10 corresponding to an embodiment. Here, a server 101, which is an information processing device that provides a currency hedging service corresponding to the embodiment, a user terminal 102, which is an information processing device used by a user receiving the currency hedging service, and a financial instruments business operator's financial transaction system 103, which performs financial transactions in response to instructions from the user terminal 102, are connected via a network 104 such as the Internet. The user terminal 102 may be configured as a general-purpose personal computer, a smartphone, or the like. Although FIG. 1 shows only one user terminal 102, this is merely for illustrative purposes; multiple user terminals 102 may be connected to the network 104, and multiple different users may receive currency hedging services from the server 101 via the user terminals 102.

[0012] In this embodiment, the user of the user terminal 102 may be, for example, a trading company that imports goods from abroad for sale in Japan or exports goods from Japan for sale abroad. In such foreign trade, when payments are made in foreign currency, the payment must be scheduled for several months (e.g., three months) from now. In such cases, exchange rates typically fluctuate over the three months from now. For example, if the yen appreciates, the yen-denominated payment amount will be smaller, so there is no problem. However, if the yen depreciates, the yen-denominated payment amount will increase, potentially reducing profits. In such cases, by making a forward exchange contract at a rate that allows profits to be confirmed, it is possible to prevent disadvantages caused by excessive exchange rate fluctuations in the yen's depreciation direction.

[0013] For example, suppose the current exchange rate (market rate) is 146 yen to the dollar. If you plan to pay $3,000,000 for a product three months from now, the current payment amount converted to yen is 438,000,000 yen. If the exchange rate strengthens, say to 144 yen at the time of settlement three months from now, the payment amount will be 432,000,000 yen, a reduction of 6,000,000 yen. In this case, if the yen strengthens, the exchange rate fluctuation benefits the user. On the other hand, if the exchange rate weakens, say to 149 yen at the time of settlement three months from now, the payment amount will be 447,000,000 yen, and the user will have to pay an extra 9,000,000 yen. If this 9,000,000 yen amount exceeds or is equivalent to the originally expected profit, it will be a significant loss to the user, and this situation must be avoided.

[0014] Therefore, in this embodiment, the amount and probability of occurrence of a limit order foreign exchange reservation for profit lock-in in the event of a stronger yen are presented. Also, the amount and probability of occurrence of a stop order foreign exchange reservation in the event of a weaker yen are presented.

[0015] In this way, it is possible to provide information for determining the profit-taking (TP: Take Profit) price in the event that the exchange rate fluctuates in the direction of a stronger yen during the period from the present time until payment, as well as information for determining the stop price in the event that the exchange rate fluctuates in the direction of a weaker yen.

[0016] After checking this information, the user can use user terminal 102 to specify the period until payment and place a limit order to buy dollars at a target rate in the direction of a stronger yen, as well as a stop-loss order to buy dollars at a defensive rate in the direction of a weaker yen, to the financial institution's gold trader system 103. This allows users to buy dollars at a limit order and lock in profits if the yen strengthens, while even if the yen weakens, buying dollars at a stop-loss order makes it possible to keep losses within expectations and avoid unexpected losses.

[0017] While the above explanation uses an importer as an example, the same applies to exporters. In the case of an exporter, for example, goods produced in Japan are exported overseas for sale overseas. Here, too, it is assumed that payment will be made in a foreign currency, and the foreign currency is currently scheduled to be received several months later (e.g., three months later). In this case, if the yen appreciates, for example, the amount received in yen terms will decrease, resulting in a decrease in profits. Conversely, if the yen weakens, the amount received in yen terms will increase, resulting in an increase in profits. In other words, importers, unlike exporters, make profits when the yen weakens and incur losses when the yen strengthens. Therefore, the handling of limit orders and stop-loss orders is reversed from that of exporters. That is, limit orders specify a target rate in the direction of a weaker yen, while stop-loss orders specify a defensive rate in the direction of a stronger yen. The system then displays the amount and probability of a limit order foreign exchange contract to lock in profits if the yen weakens. The system also displays the amount and probability of a stop-loss foreign exchange contract to lock in profits if the yen weakens. In this way, even in the case of importers, by providing information for making exchange reservations within a range where profits can be confirmed, it is possible to prevent disadvantages caused by the exchange rate fluctuating more than necessary in the direction of a stronger yen.

[0018] In this way, in this embodiment, the occurrence rates (also called realization probability or achievement probability; the same applies below) of the target rate and defense rate are presented for the planned payment amount and the planned receipt amount, thereby providing the user with information regarding the feasibility (or certainty) of realizing profits and avoiding losses.

[0019] <Hardware configuration> Next, the hardware configuration of the server 101 will be described with reference to Fig. 2. The server 101 can be configured as an information processing device, for example, from one or more personal computers. In Fig. 2, a CPU 201 as a processor controls the server 101 using programs and data stored in a RAM (random access memory) 202 and a ROM (read only memory) 205 as memories, and executes processes corresponding to the embodiments described below. The RAM 202 has an area for reading processing programs stored in an internal storage device 207 and information stored in an external storage device 208, and also has a work area used by the CPU 201 when executing various processes. The server 101 may use a cloud service.

[0020] The input unit 203 is an input means for receiving input from the administrator of the server 101, and is composed of a keyboard, a mouse, etc. The communication I / F (interface) 204 functions as an I / F for connecting to the network 106, etc. The ROM 205 stores programs (such as a boot program) that control the entire server 101, etc. The display unit 206 is a display unit that serves as a display screen, and is composed of a liquid crystal display device, etc.

[0021] The internal storage device 207 is mainly composed of a hard disk, and stores programs and various application data for the server 101 to execute processes. The data stored here is read out to the RAM 202 as needed. The external storage device 208 is a database, and stores user information, exchange rate information, etc. as needed. The bus 209 provides interconnection between the above-mentioned blocks.

[0022] FIG. 2 has been described as the hardware configuration of the server 101, but the basic hardware configuration of the user terminal 102 can also be the same as that shown in FIG.

[0023] <Processing flow in System 10> Next, a description will be given of the processing executed in the system 10. Fig. 3 is a timing chart showing an outline of the flow of processing executed in the system 10 corresponding to this embodiment. The processing is realized by the CPUs of the server 110 and the user terminal 102 executing corresponding processing programs, respectively, or by the CPU controlling the operation of functional blocks such as the input unit 203, communication I / F 204, and display unit 206 shown in Fig. 2.

[0024] First, in S301 of FIG. 3, the user terminal 102 displays a login screen and accepts input of login information. In S302, the accepted login information is transmitted from the user terminal to the server 101. In S303, the server 101 authenticates the user and, if the authentication is successful, acquires user information registered for the user. Subsequently, in S304, the server 101 transmits main screen information to the user terminal 102. In S305, the user terminal 102 displays the main screen on the display and accepts input from the user. In S307, the accepted input information is transmitted from the user terminal to the server 101. In S308, the server 101 performs processing based on the contents of the received input information and generates screen information. The generated screen information is transmitted from the server 101 to the user terminal 102 in S309.

[0025] The user terminal 102 displays the received screen information on the display in S310 and accepts input from the user in S311. The accepted input information is transmitted from the user terminal 102 to the server 101 in S312. The server 101 executes processing in the same manner as in S308 in accordance with the transmitted input information, generates screen information, and transmits it back to the user terminal 102.

[0026] <Main screen> The processes from S305 to S312 can be repeated between the user terminal 102 and the server 101. The flow of the processes will be explained with reference to the screens in FIG. 4 and subsequent figures. FIG. 4A is a diagram showing an example of a main screen corresponding to this embodiment. In S305, this screen is displayed on the user terminal 102.

[0027] Various pieces of information are displayed on screen 400. First, a menu display area 401 is displayed. This menu display area 401 displays information items that can be displayed on screen 400, and the user can select the information item to display from menu display area 401. FIG. 4A shows a state in which the dashboard has been selected. Other information items that can be displayed may include a hedge guide, foreign currency management, exchange rate hedge effects, account management, logout, etc.

[0028] Screen 400, as a main screen showing a dashboard, displays information 402 for the following items at the top: hedge ratio, lower limit alert, and upper limit alert. The hedge ratio indicates the ratio (application rate) at which the amount of foreign currency hedging is calculated by applying the service provided by server 101 (referred to as the "hedge service"), and is set to 100% in FIG. 4A. This means that, for example, if a foreign currency payment of $10,000 is planned, 100% of the $10,000, or the entire $10,000, is subject to the hedge service provided by server 101. If the hedge ratio is set to 50%, then $5,000, or 50% of the $10,000, is subject to the hedge service. In this embodiment, an alert is set to be sent to the user when the prevailing exchange rate touches the upper or lower limit, and the lower limit alert and upper limit alert are set to respective amounts. In FIG. 4A, the lower limit is 130 yen and the upper limit is 160 yen.

[0029] 4A, the upward direction is the direction in which the value of the Japanese yen increases (or the positive direction), and the downward direction is the direction in which the value of the Japanese yen decreases (the negative direction). Generally, the value of the Japanese yen against the US dollar decreases as the value of the Japanese yen increases (yen depreciation), and increases as the value decreases (yen appreciation). However, the concepts of upward (positive direction) and downward (negative direction) in this embodiment are separate from the value of the Japanese yen against the US dollar and refer simply to the increase or decrease of the amount.

[0030] In this embodiment, we will mainly explain the case of paying US dollars from a US dollar foreign currency account among payment and receipt transactions at the exchange rate between Japanese yen and US dollars. In addition to US dollars, foreign currencies such as euros, pounds, Swiss francs, Australian dollars, Canadian dollars, Singapore dollars, Chinese yuan, Indian rupees, Indonesian ringgits, Thai baht, Philippine pesos, South Korean won, and Vietnamese dong can also be used.

[0031] The trend forecast graph 403 shows the trend (transition) of the exchange rate (dollar / yen) for approximately one year from the date the screen is displayed (today) as the base date, and predicts exchange rate trends from today onward with a certain range. For the past one year, the actual exchange rate movement is shown as a single broken line on the left side of today. Meanwhile, the right side shows exchange rate trends predicted by artificial intelligence (AI) using multiple broken lines or curves. In this embodiment, N (N is, for example, 100, 200, 300, etc.) AIs predict exchange rate trends for a predetermined number of years into the future (e.g., one year, two years, five years, etc.). All predictions are sorted from those predicting a decline in the value of the Japanese yen (yen appreciation) to those predicting an increase in the value (yen depreciation). The first and last predictions, as well as predictions in 10% increments from the top 10% to the top 90%, are displayed in the trend forecast graph 403 as representative exchange rate predictions for the rest. For example, if there are 100 AIs, the predictions ranked 1st, 10th, 20th, 90th, and 100th from the top can be selected as representatives.

[0032] In FIG. 4A, the 50th value is indicated by a dashed line. If the 50th value is lower than today's actual market rate (146.93 yen), this indicates that a large percentage of people predict that the future trend will be toward a stronger yen. On the other hand, if the 50th value is higher than today's actual market rate (146.93 yen), i.e., the yen is depreciating, this indicates that a large percentage of people predict that the future trend will be toward a weaker yen. The trend forecast graph 403 can be displayed with the cursor placed on it to display the forecast value for any position (year, month, and day). In addition, the lower limit alert and upper limit alert are indicated by thick dotted lines.

[0033] Display 404 shows the current rate at the current time, and this value becomes the value for today in trend prediction graph 403. Screen 400 also includes AI predicted distribution 405 and AI voting percentage 406.

[0034] The AI ​​forecast distribution 405 displays the upper and lower limits of each forecast rate at any point in time, for example, predicted by 100 AIs. Here, the arbitrary points in time are shown as 1 month, 3 months, 6 months, 9 months, and 12 months later, but are not limited to these. The upper and lower limits are the upper and lower limits of the forecast rates by the remaining 80% of the AIs that made the predictions, excluding the top 10% and bottom 10% of the AIs that made the predictions. The display of the AI ​​forecast distribution 405 allows one to read the overall trend of the forecasted exchange rate trends. The AI ​​forecast distribution 405 may also display the average value of the forecast rate at any point in time.

[0035] Next, the AI voting ratio 406 indicates, for each prediction rate at any given time predicted by, for example, 100 AIs, whether the prediction is in the direction of yen depreciation or yen appreciation, and shows the ratio. For example, if it is one month later, it can be seen that 40% of the total AIs predict yen depreciation and 60% of the total AIs predict yen appreciation. In addition to displaying the ratio as a number, the ratio may also be shown by a bar graph or the like. In this embodiment, the ratio of the prediction in the yen depreciation direction is called the first ratio, and the ratio of the prediction in the yen appreciation direction is called the second ratio. By referring to these ratios, the future trend of the exchange rate can be grasped.

[0036] <AI Model Generation Method> In this embodiment, a plurality of AI models are used to predict the future transition of the exchange rate. Hereinafter, the method for generating the AI model in this embodiment will be described.

[0037] In this embodiment, an AI model for exchange rate trend prediction is generated by deep learning more than 400 types of information suitable for medium- and long-term prediction, including the following teacher data. The teacher data includes, for example, historical price information, technical indicators, Japan-US policy interest rates, dot charts (future interest rate outlook information of FOMC members), IMM currency futures positions, gross domestic product, consumer price index, employment statistics, Japan-US trade balance, Nikkei average stock price · S&P 500, crude oil · gold · copper prices, factor decomposition of market fluctuations by stochastic differential equations, etc.

[0038] Using such teacher data, AI models with slightly different personalities are generated through model learning based on deep learning and Transformer technology. For example, during learning, if the application order of the above-mentioned teacher data changes, naturally different AI models will be generated. In this way, by using a large number of AI models with slightly different personalities or strengths, even in a market where a certain AI model cannot predict well, other AI models can complement the error and finally enable accurate prediction. The future trend prediction by the AI model can be performed, for example, by time series prediction using deep learning technology. Thereby, it is possible to tackle the "recovery of hidden clues" behind the market.

[0039] Although the trend forecast graph 403 only shows the forecasts of some AI models, in reality, forecasts from many AI models are generated for a certain period in the future, and for each forecast, it is possible to determine whether a specific exchange rate will be realized within a specific period. For example, when a forward exchange contract is to be executed with a limit order of 143 yen and a due date three months from now against a current prevailing rate of 145 yen, it is possible to determine for each of the forecasts of multiple AI models whether 143 yen will be realized within the next three months. For example, if there are 100 AI models and 40 of them predict that the forward exchange rate will be realized, the probability of realization is 40%. Similarly, when a forward exchange contract is to be executed with a stop order of 150 yen and a due date three months from now, it is possible to determine for each of the forecasts of multiple AI models whether 150 yen will be realized within the next three months. For example, if 44 of the 100 AI models predict that the forward exchange rate will be realized, the probability of realization is 44%.

[0040] In this way, in this embodiment, it is possible to predict future exchange rate trends and the feasibility of a specific exchange rate over a certain period of time from the overall trend of predictions from many AI models with different personalities, rather than relying on the predictions of any individual AI model. As the number of AI models increases, the reliability of the overall trend and feasibility increases.

[0041] After displaying a screen such as that shown in FIG. 4A in S305 of FIG. 3, the user terminal 102 accepts an operation from the user in S306 and transmits the accepted operation content to the server 101 as input information in S307. The user operation includes, for example, selecting a menu item in the menu display area 401. If the user operation requests the display of a new screen, the server 101 executes processing required to generate the screen in S308 and transmits the processing result to the user terminal 102 as screen information in S309. For example, if "Rate Defense Guide" is selected in the menu display area 401, the server 101 generates screen information such as that shown in FIG. 4B and transmits it to the user terminal 102. The user terminal 102 then displays the received screen information in S310, accepts further input from the user in S311, and transmits the input information to the server 101 in S312. The server 101 processes the input information in S308 in the same way as the input information in S307, and transmits the processing result to the user terminal 102. After this, the processes from S310 to S312 and S307 to S309 are repeated.

[0042] <Rate Defense Guide> In this embodiment, a rate defense guide can be provided as a hedging guide to keep the transaction rate within a certain range toward the due date for payment or receipt of foreign currency. The rate defense guide presents information on the feasibility of specified limit and stop-loss rates for the transaction rate of the amount of foreign currency on the due date for payment or delivery. In this embodiment, a limit order refers to an order method for buying below a specified price or selling above a specified price, such as "buy below XX yen" or "sell above XX yen." A stop-loss order, the opposite of a limit order, refers to an order method for buying above a specified price or selling below a specified price, such as "buy above XX yen" or "sell below XX yen." Thus, for example, a limit order rate of 143 yen means buying when the yen strengthens to 143 yen or below the dollar, and a stop-loss order rate of 155 yen means buying when the yen weakens to 155 yen or above the dollar. This applies to foreign currency payments, but the opposite applies to foreign currency receipts. In other words, a limit rate of 155 yen means selling if the exchange rate moves in the direction of a weaker yen, to 155 yen per dollar or more, and a stop-reverse rate of 143 yen means selling if the exchange rate moves in the direction of a stronger yen, to 143 yen per dollar or less. Users can set any amount for the limit rate and stop-loss rate for foreign currency payments or receipts. However, for payment, the stop-loss rate must be higher than the current exchange rate (equivalent to the rate when the yen weakens), and for receipt, the stop-loss rate must be lower than the current exchange rate (equivalent to the rate when the yen strengthens).

[0043] <Foreign currency hedging guide settings> FIG. 4B shows an example of a screen displayed when "Rate Defense Guide" is selected from "Hedge Guide" in the menu display area 401. The rate defense guide screen 410 displays a display 411 indicating that the current screen is the "1. Currency Hedge Guide Setting" screen and that the target exchange rate is the U.S. dollar - Japanese yen (USDJPY). The hedge ratio setting interface 412 allows the user to set the hedge ratio to any value between 0% and 100%. The hedge ratio indicates the percentage of the payment amount or receipt amount to be hedged for rate defense. At 100%, the entire payment amount or receipt amount (or the difference between them) becomes the hedged amount, and the hedged amount is subtracted from this to create the rate defense guide. While FIG. 4B shows a 100% setting as an example, a lower application rate may also be used. In FIG. 4B, the user can enter lower and upper limit alerts using input fields 413a and 413b, respectively. The lower and upper limit amounts entered here are reflected at the dotted line positions in the trend forecast graph 403 shown below. The operations for setting the application rate, lower limit alert, and upper limit alert above correspond to the operations accepted by the user terminal in S306.

[0044] In Figure 4B, below the trend forecast graph 403, there are displayed input areas for setting a hedge guide notification date 414, setting bank fees 415, forward exchange quota 416, minimum recommended additional hedge amount 417, and long-term exchange setting 418. In these areas, it is possible to select items and input values, which corresponds to the operations accepted by the user terminal in S306 (or S311).

[0045] Here, the desired date for the notification of the hedge guide can be set in the hedge guide notification date setting 414. In this embodiment, it is recommended to create a foreign exchange hedge guide once a month, and a notification encouraging the use of the hedge guide is sent on the date set in the hedge guide notification date setting 414. In the bank fee setting 415, the exchange fee (spread) set by the bank for the customer is input. The value input here is used to calculate the reservation rate displayed in the hedge guide in Figure 4F. The reservation rate is calculated using the following formula 1. Forward Rate = Interbank Offer Rate + Bank Fee (Equation 1)

[0046] The forward exchange limit 416 indicates the total amount of forward exchange contracts permitted by the bank. For example, if the user's forward exchange limit is $1 million, even if the hedging guide recommends a hedge transaction of more than $1 million, only $1 million can be traded. The minimum recommended additional hedge amount 417 indicates the minimum amount for displaying the additional hedge amount. For example, if the minimum recommended additional hedge amount is $100,000, the recommended hedge amount is $300,000, and the hedged amount is $250,000, the difference of $50,000 is less than the minimum recommended additional hedge amount of $100,000 ($100,000 > $300,000 - $250,000), so the recommended additional hedge amount must be set to 0, or the user is prompted to re-enter the amount. Note that the minimum amount may be applied to the absolute value of the difference, or may be applied only to positive difference values. When applied to the absolute value, even if the hedged amount is larger and the sign of the difference value is negative, if the absolute value exceeds the minimum amount, the user can still hedge the amount by selling. Also, when applying only to positive differential values, even if the absolute value exceeds the minimum amount, re-entry is prompted if the sign of the differential value is negative. In the long-term exchange setting 418, it is possible to set whether to consider long-term exchange extinction conditions such as coupon swaps when creating a rate defense guide. If the setting is to consider extinction conditions, the guide is created assuming that the current prevailing rate will continue into the future and that foreign currency deliveries via long-term exchange will cease after the year and month in which the long-term exchange meets the extinction conditions. If the setting is to not consider extinction conditions, the guide is created assuming that foreign currency deliveries will occur periodically during the long-term exchange delivery period, regardless of the extinction conditions.

[0047] When this information is entered and the “Next” button 419 is operated, the entered information is sent to the server 101, and the screen information shown in FIG. 4C is generated in the server 101 and sent to the user terminal 102.

[0048] <Foreign currency payment and receipt plan> First, in FIG. 4C, display 421 indicates that the current screen is the "2. Input Foreign Currency Payment and Receipt Plan" screen. Screen 420 also displays a bar graph display area 422 displaying the monthly total and hedged amounts of foreign currency payment and delivery. Bar graph 422A shows the foreign currency payment and delivery amount for October 2024. Bar graph 422B shows the hedged amount for the period due in October 2024. While not shown in FIG. 4C because the hedge ratio is 100%, if the hedge ratio is less than 100%, e.g., 80%, a marker indicating the 80% point on bar graph 422A may be displayed. For example, the color of bar graph 422A may be changed at the 80% point, or a line may be displayed at the 80% point. This allows for a visual understanding of the size of the hedged amount relative to the payment amount, etc. Below the bar graph display area 422, a table 423 is displayed for registering the monthly foreign currency payment amount, foreign currency delivery amount, hedged amount, and average hedge rate. Table 423 includes check boxes, and the following items: foreign currency payment receipt plan - net, foreign currency payment plan, foreign currency receipt plan, hedged amount, and average hedge rate. Of these, numerical values ​​can be entered for the foreign currency payment plan and foreign currency receipt plan. The foreign currency payment plan is an input field for entering the foreign currency payment amount scheduled to be paid that month. The foreign currency receipt plan is an input field for entering the foreign currency receipt amount scheduled to be received that month. Numerical values ​​are entered into these input fields on a monthly basis. The amount of the foreign currency receipt plan - net and the display of bar graph 422 are also updated according to the values ​​entered.

[0049] The hedged amount and average hedge rate can be entered on a separate screen, but may also be entered on screen 420. The hedged amount can be entered by selecting "Foreign Currency Management" and "Exchange Hedge Transaction Management" in menu display area 401, and then entering executed hedge information on the screen displayed. When entering the hedged amount, the contract date, currency pair (e.g., USD / JPY), forward exchange contract type (fixed date delivery, term delivery, coupon swap), delivery date, rate, receipt amount, payment amount, etc. can be specified. The rate, delivery date, receipt amount, and payment amount specified here are used to display table 423. That is, the hedged amount is first displayed for the month to which the delivery date belongs. Furthermore, the rate information for hedges with delivery dates arriving in that month is averaged and displayed as the average hedge rate. Furthermore, for the hedged amount, for example, in the case of an importer, the payment amount is displayed as a positive value and the receipt amount is displayed as a negative value, resulting in a total.

[0050] In Figure 4C, a payment of $3,000,000 is scheduled for October, and the receivable amount is zero, resulting in a net foreign currency payment and receipt plan of $3,000,000. If a receivable amount were entered, e.g., if a receipt of $2,000,000 were entered, the net foreign currency payment and receipt plan would be the difference of $1,000,000. Similarly, if a receipt of $4,000,000 were entered, the net foreign currency payment and receipt plan would be the difference of -$1,000,000. A checkbox is checked to select the month to be covered by the rate defense guide. Selecting the checkbox corresponds to specifying the maturity date of the forward exchange contract that serves as the basis for creating the rate defense guide. In this embodiment, a monthly blanket hedge is assumed, so the concept of "maturity" does not refer to a specific date, but rather to a year and month. Although Figure 4C does not show the date only as the year and month, the due date may be the last day of the checked month, the first or middle day of the month, or any date set arbitrarily.

[0051] Also, in Figure 4C, a $1,500,000 hedged purchase with a specified October receipt date has already been executed, so this amount is subtracted from the hedged amount. If the hedged amount is zero, the amount of the foreign currency payment receipt schedule - net becomes the amount to be used for creating the guide. Also, if the hedged amount is greater than the foreign currency payment receipt schedule - net amount (for example, if the foreign currency payment receipt schedule - net is a $3,000,000 payment and the hedged amount is a $4,000,000 purchase), the purchase amount exceeds the amount of $1,000,000. In this case, the required amount has already been hedged, so the additional hedge amount in the rate defense guide is set to zero. If the difference between the foreign currency payment receipt schedule - net and the hedged amount is smaller than the amount specified in the recommended additional hedge minimum amount 417 in Figure 4B, the recommended additional hedge amount can be set to zero, or the user can be prompted to re-enter the foreign currency payment schedule and foreign currency receipt schedule.

[0052] In Figure 4C, only October 2024 is checked, so a guide will be generated for a payment of $1,500,000. If a rate defense guide is desired for the receipt and payment amounts for other months, the guide will be created for those months by checking the checkboxes. These item selections and numerical inputs correspond to the operations accepted by the user terminal in S306 (or S311). After completing the input of the foreign currency receipt plan and foreign currency payment plan, when the "Next" button 424 is operated, the input information is sent to the server 101, and the screen information shown in Figure 4D is generated in the server 101 and sent to the user terminal 102.

[0053] <Rate Defense Guide Settings> In Figure 4D, a screen 430 for inputting a desired rate is displayed. Screen 430 indicates, by display 431, that the current screen is the "3. Rate Defense Guide Setting Input" screen. Below this, the same trend prediction graph 403 as that displayed in Figure 4B is displayed, and furthermore, a target rate setting field 432 and a defense rate setting field 433 for inputting two types of desired rates are displayed.

[0054] The user can input either or both of a first desired rate equivalent to a limit order and a second desired rate equivalent to a stop order in the target rate setting field 432 and the defense rate setting field 433. This is to meet the needs of users, as their circumstances vary: some users are only interested in taking profits at limit orders, some are only interested in stop losses, and some are interested in both taking profits and stop losses.

[0055] Although the relationship between limit orders and stop orders is reversed for foreign currency payments and foreign currency receipts, in this embodiment, they can be determined based on the sign of the net foreign currency payment and receipt plan. That is, if the net foreign currency payment and receipt plan has a positive sign, payments are favored, so the target rate is a rate in the direction of a stronger yen, and the stop order is a rate in the direction of a weaker yen. On the other hand, if the net foreign currency payment and receipt plan has a negative sign, receipts are favored, so the target rate is a rate in the direction of a weaker yen, and the stop order is a rate in the direction of a stronger yen. Separately, whether payment or receipt should be prioritized may be preset based on the type of business of the user. For example, foreign currency payments may be prioritized for users who engage in import business, while foreign currency receipts may be prioritized for users who engage in export business. For users who engage in both import and export, the target rate (first desired rate) and defense rate (second desired rate) can be determined based on the sign of the net foreign currency payment and receipt plan, as described above.

[0056] FIG. 4D assumes that a rate defense guide is being created for a payment of $1,500,000 excluding the hedged amount based on the input results in FIG. 4C. Therefore, the target rate is set toward a stronger yen. That is, the objective is to convert $1,500,000 to yen and secure a foreign exchange gain when the yen strengthens from the current rate and reaches the target rate. However, it is not necessarily required to set the yen stronger than the current exchange rate. On the other hand, the defense rate is set toward a weaker yen. Therefore, an amount greater than the current exchange rate is required to be input. In this embodiment, the current exchange rate is $146.93, so an amount greater than this is input. The objective is to convert $1,500,000 to yen and cut losses when the yen weakens from the current rate and reaches the defense rate.

[0057] Although not shown in FIG. 4D , when creating a rate defense guide for receiving $1,500,000, the target rate is set toward a weaker yen. That is, the objective is to sell $1,500,000 and convert it to yen to secure a foreign exchange gain when the yen weakens from the current rate and reaches the target rate. However, it is not necessarily required to set the yen weaker than the current exchange rate. On the other hand, the defense rate is set toward a stronger yen. Therefore, an amount smaller than the current exchange rate is required. In this embodiment, since the current exchange rate is $146.93, an amount smaller than this is input. Then, when the yen strengthens from the current rate and reaches the defense rate, the $1,500,000 is converted to yen and a loss is cut. The target rate and defense rate are each displayed as a two-dot chain line in the trend forecast graph 403.

[0058] In Figure 4C, if amounts are entered for both the foreign currency payment schedule and the foreign currency receipt schedule, the amount for which the rate defense guide is generated is the difference between the two amounts. For example, if the foreign currency payment schedule is $3,000,000 and the foreign currency receipt schedule is $1,200,000, the net foreign currency payment receipt schedule will be a payment of $1,800,000. This means that the $1,200,000 in foreign currency receipts can be directly applied to foreign currency payments, eliminating the need for forward exchange contracts, and a rate defense guide will be generated for the excess payment of $1,800,000. The same applies if the sign of the net foreign currency payment receipt schedule is negative; a rate defense guide will be generated for the excess receipt. However, if there is a hedged amount, that amount will be taken into account as described above.

[0059] When a target rate as a limit order and a defensive rate as a stop-loss order are specified on screen 430 of FIG. 4D and the “Create guide” button 434 is operated, the entered information is sent to server 101, and the screen information shown in FIG. 4E is generated in server 101 and sent to user terminal 102.

[0060] <Hedge Guide Results> The display screen for the rate defense guide results will be described with reference to Figure 4E. Screen 440 indicates in display 441 that the current screen is the "4 Rate Defense Guide Results" screen. Screen 440 displays the current exchange rate 442, which is the current exchange rate at the time the rate defense guide was generated. Display area 443 displays exchange reservation information, and the difference between the amount secured as the reservation limit and the amount currently reserved is displayed as the available reservation amount.

[0061] In the display area 444, the additional hedge amount is displayed together with the forward rate, as well as the expiration date, limit rate, and stop rate. The expiration date is shown in months. The forward rate is merely a reference exchange rate and is calculated based on the above formula 1. The limit rate corresponds to the amount specified in the target rate 432 and is calculated based on the following formula 2. Limit rate = (Target rate × Foreign currency payment plan - Net × Hedge ratio - Average hedge rate × Hedged amount) / (Foreign currency payment plan - Net × Hedge ratio - Hedged amount) (Equation 2) For example, if the target rate is 143 yen, the foreign currency payment and receipt plan - net is $3,000,000, the average hedge rate is 146, the hedged amount is $1,500,000, and the hedge ratio is 100%, then by substituting these values ​​into formula 2, the target rate is (143 x 3,000,000 - 146 x 1,500,000) / 1,500,000 = 140 yen. If the hedge ratio is, for example, 80%, the target rate would be 138 yen by performing the same calculation.

[0062] In addition, the stop-loss rate corresponds to the amount specified in the defense rate 433 and is calculated based on the following formula 3. Stop-loss rate = (Defense rate × Foreign currency payment and receipt plan - Net × Hedge ratio - Average hedge rate × Hedged amount) / (Foreign currency payment and receipt plan - Net × Hedge ratio - Hedged amount) (Equation 3) For example, if the defense rate is 150 yen, the foreign currency payment and receipt plan - net is $3,000,000, the average hedge rate is 146, and the hedged amount is $1,500,000, then by substituting these values ​​into formula 2, the target rate is (150 x 3,000,000 - 146 x 1,500,000) / 1,500,000 = 154 yen. If the hedge ratio is 80%, then the target rate is calculated in the same way and is 156.67 yen.

[0063] In the above, since there is a hedged amount, the target rate and limit rate do not match, and the defense rate and stop rate do not match, but if the hedged amount is 0, the target rate and limit rate match, and the defense rate and stop rate match. In addition, display area 445 displays the hedged amount, as well as the hedged amount and the unhedged amount. As mentioned above, if there is a hedged amount, it is displayed as the hedged amount, and that amount is subtracted from the additional hedged amount. The hedged amount is the foreign currency receipt plan - net amount for the month for which the checkbox was checked in Figure 4C. Since only October 2024 was selected in Figure 4C, only October 2024 is shown as the due date in Figure 4E, but if other months are selected, the additional hedged amount, etc. will be displayed for each due date.

[0064] Furthermore, because the hedge ratio setting 412 in Figure 4B is set to 100%, the entire amount of the foreign currency receipt plan - net is subject to the calculation of the limit rate and stop-loss rate. For example, if the hedge ratio setting 412 is set to 80%, the hedged amount is determined by subtracting the hedged amount from $2,400,000. In the example above, $1,500,000 is hedged, so the hedged amount is $900,000 excluding the hedged amount. It is recommended that a foreign exchange contract be made for this hedged amount. The exchange rates for this may be specified as the limit price indicated by the limit rate and the stop-loss price indicated by the stop-loss rate. However, foreign exchange contracts can only be made within the range of amounts indicated in the foreign exchange contract information 443.

[0065] In FIG. 4E, the occurrence rates (occurrence probability, realization probability) are displayed for the limit rate and the stop rate. The limit rate is 7.81% for 140 yen. The stop rate is 11.74% for 154 yen. These occurrence rates can be calculated based on the prediction of future exchange rate trends using multiple AI models, as described above. In this embodiment, the multiple AI models predict exchange rate trends from today onward within a certain range. For example, for a limit rate of 140 yen, the occurrence rate for the limit rate of 140 yen is calculated based on the ratio of the number of AI models that predict the yen will appreciate beyond 140 yen by the end of October 2024 to the total number of AI models. In this case, the ratio of AI models that predict the yen will appreciate beyond 140 yen by the end of October 2024 is 7.81%. On the other hand, for the stop-loss rate of 154 yen, the occurrence rate of the limit rate of 154 yen is calculated based on the ratio of the number of AI models that predict that the yen will weaken to less than 154 yen in the period up to the end of October 2024 to the total number of AI models. In this case, the percentage of AI models that predict that the yen will weaken to less than 154 yen in the period up to the end of October 2024 is 11.74%. Note that the end date of the period for calculating the occurrence rate is not limited to the end of the month. For example, it may be the beginning of the month (the 1st), or a specific date such as the 15th or 25th. Furthermore, the user may be able to specify it at their own discretion.

[0066] In this way, by checking the occurrence rates of the limit order rate and stop-loss rate, it is possible to predict their feasibility. Therefore, the user can return to the screen of FIG. 4D, re-enter the target rate and defense rate, and re-generate the guide, thereby checking the occurrence rates of the re-entered target rate and defense rate. This allows the user to check the occurrence rates and adjust the exchange rate at which they wish to take profits or cut losses.

[0067] Although it is sufficient to make a forward exchange contract for the rate defense guide once, if the hedge ratio setting 412 is set low, it may be necessary to determine whether additional hedging is necessary for the unhedged portion. In the above example, a notice regarding additional hedging is sent on the date set in the hedge guide notice date setting 414 in Figure 4B.

[0068] The PDF output 446 button in Fig. 4E is an operation button for outputting the contents of screen 440 as PDF data. Since the exchange reservation itself must be processed separately from this interface, it is possible to refer to the PDF and execute it at that time.

[0069] Figure 4E above shows an example of a guide screen created when the foreign currency receipt plan - net is positive and the target rate 432 on screen 430 of Figure 4D is lower than the current exchange rate (market rate) (the amount by which the yen has strengthened). In contrast, if the foreign currency receipt plan - net is positive and the target rate 432 on screen 430 of Figure 4D is higher than the current exchange rate (market rate) (the amount by which the yen has weakened), for example, if 148 yen is entered as target rate 432 against a market rate of 146.93 yen / dollar, the content of the guide screen created will be different. Figure 4F shows an example of a guide screen created in this case.

[0070] The example of the screen display shown in FIG. 4F is basically the same as that shown in FIG. 4E. However, instead of display area 444, display area 451 is displayed. Like display area 444, display area 451 displays the due date, forward rate, and additional hedge amount, but does not display the limit rate, stop rate, or their occurrence rates. Instead, a notification message is displayed for the user. The notification message recommends that it is better to make an exchange contract at the prevailing rate rather than making an exchange contract at the target rate amount entered on the screen of FIG. 4D. FIG. 4F displays the message, "At the current prevailing rate, if you hedge the additional hedge amount with an exchange contract, you may be able to pay at a rate more favorable than the target rate." However, the content of the message is not limited to this text.

[0071] The screen display in Fig. 4F is displayed in the case where the sign of the foreign currency receipt plan - Net is negative, and the target rate 432 on the screen 430 in Fig. 4D is lower (the amount has moved in the direction of a stronger yen) than the current exchange rate (market rate), for example, when the market rate is 146.93 yen / dollar and 145 yen is entered as the target rate 432. The above describes the case where a comparison with the market rate is made, but a screen like that in Fig. 4F may also be displayed when comparing with a forward rate.

[0072] In this embodiment, the additional hedging amount is presented in display areas 444 and 451, but the amount that can actually be hedged is the reservable amount shown in display area 443, so the user can either secure a reservable amount that will allow them to secure the additional hedging amount, or return to the screen in Figure 4C and reselect the hedging target, or adjust the hedge ratio setting 412 in Figure 4B to lower the hedge ratio, so that the additional hedging amount falls within the reservable amount.

[0073] <Effects of the embodiment> As described above, according to this embodiment, foreign currency transaction rates can be kept within a certain range by making an exchange reservation for the amount of foreign currency to be paid, the amount to be received, or the difference between them, on a future date currently expected to be paid or received, so that profits are locked in or losses are cut when a certain exchange rate is reached in accordance with a prediction of future exchange rate fluctuations.

[0074] Specifically, in the dollar-yen exchange rate, the future exchange rate will move either in the direction of a stronger or weaker yen, and it is difficult to determine at this point which way it will move on the maturity date. However, if you leave it until the maturity date and make a payment (receive) for dollars at the spot price on that date, the yen may weaken (appreciate) more than it is now, resulting in more Japanese yen being paid (less Japanese yen being received), and the exchange rate may ultimately exceed the break-even point. Furthermore, even if the yen were to strengthen (weaken) significantly by the maturity date, you would not be able to buy (sell) dollars at that time, meaning you would miss the opportunity to lock in your profits. On the other hand, even if you enter into a forward exchange contract, betting on a stronger (weaker) yen that is unlikely to materialize will still result in you missing your opportunity to lock in your profits, and overestimating your stop-loss exchange rate too low (high) could also result in you missing your opportunity to lock in your profits.

[0075] In response to this, by using multiple AI models to predict future trends and presenting the probability of a specific exchange rate being realized in the period leading up to the expiry date, it becomes possible to set limit and stop-loss exchange rates so as not to miss the opportunity to lock in profits.

[0076] Furthermore, if the yen is predicted to weaken over the period leading up to the expiration date and the rate specified in the limit order is unlikely to be realized, it is anticipated that it would be more advantageous to buy dollars at the current spot price when it comes to making dollar payments, so in such cases the system can recommend that the user buy dollars at the current time. Conversely, if the yen is predicted to strengthen when it comes to receiving dollars and the rate specified in the limit order is unlikely to be realized, it is anticipated that it would be more advantageous to sell dollars at the current spot price, so in such cases the system can also recommend that the user sell dollars at the current time. This allows the user to lock in profits.

[0077] In this way, by making forward exchange contracts with limit orders and stop orders that take feasibility into consideration, it is possible to reduce the impact of future exchange rate fluctuations while keeping the fluctuation range of payment amounts and receipt amounts within a certain range.

[0078] In the above-described embodiment, the exchange rates based on the first currency and the second currency are described only for the combination of the first currency and the second currency, i.e., the US dollar-Japanese yen or the Japanese yen-US dollar. However, the combination of the first currency and the second currency is not limited to this, and any other combination of currencies can be applied, such as the euro-Japanese yen, the British pound-Japanese yen, the euro-US dollar, the euro-British pound, the Japanese yen-Chinese yuan, etc. In addition, other currencies that can be used as combinations can include the Swiss franc, the Australian dollar, the Singapore dollar, the Indian rupee, the Indonesian ringgit, the Thai baht, the Philippine peso, the South Korean won, the Vietnamese dong, etc.

[0079] <Summary of the embodiment> The above-described embodiment discloses at least the following information processing device and computer program. (1) A server that provides information about currency hedging to a user terminal, one or more processors; Memory and a program stored in the memory that, when executed by the one or more processors, Using a plurality of AI models, obtaining, for each AI model, a forecast of future changes in an exchange rate indicating the value of a second currency relative to a first currency; accepting a designation of a first amount of the first currency from the user terminal; receiving input of at least one of a first desired exchange rate and a second desired exchange rate for the first amount of foreign exchange contract, and an input of a maturity date of the foreign exchange contract; calculating a limit price for the exchange rate for a forward foreign exchange contract of the first amount based on the received first desired rate, the first amount, and a hedged amount already hedged in relation to the maturity date; calculating a stop loss of the exchange rate for the forward exchange contract of the first amount based on the received second desired rate, the first amount, and the hedged amount; Calculating a first realization probability of the calculated limit price during the period up to the expiration date based on future trends in the exchange rate individually predicted by the plurality of AI models; Calculating a second realization probability of the calculated stop loss order during the period up to the expiration date based on future trends in the exchange rate individually predicted by the plurality of AI models; transmitting information on a second amount of the first amount to be subject to currency hedging to the user terminal together with information on the calculated limit order and the calculated first realization probability, and information on the calculated stop order and the calculated second realization probability; The server that runs the (2) the first realization probability is calculated based on a ratio of the AI ​​models among the plurality of AI models that predicted an event in which the exchange rate reaches the limit price within the period until the due date; The server described in (1), wherein the second realization probability is calculated based on the proportion of AI models among the plurality of AI models that predicted the event that the exchange rate would reach the stop-loss price within the period until the due date. (3) Receiving a designation of a first amount of the first currency from the user terminal includes: receiving, from the user terminal, a designation of a third amount of the first currency to be paid on the due date; receiving, from the user terminal, a designation of a fourth amount of the first currency to be received on the due date; Including, The server according to (2), wherein the difference between the third amount and the fourth amount is designated as the first amount. (4) if the sign of the first amount is positive, a second desired rate is specified as an amount of the exchange rate change from the current exchange rate in a first direction that devalues ​​the second currency; The server of (3), wherein if the sign of the first amount is negative, a second desired rate is specified as the amount of the exchange rate change from the current exchange rate in a second direction in which the value of the second currency increases. (5) The program, when executed by the one or more processors, causes the server to: The server described in (4) further executes the following: if the sign of the first amount is positive and the amount indicated by the limit order is higher than the amount based on the current exchange rate, or if the sign of the first amount is negative and the amount indicated by the limit order is lower than the amount based on the current exchange rate, sending to the user terminal, together with information on the second amount, a notification that an exchange reservation at the current exchange rate is recommended, instead of information on the limit order and the first realization probability and information on the stop order and the second realization probability. (6) When the program is executed by the one or more processors, the program further causes the server to receive an input specifying a proportion of the first amount to be subject to the currency hedging; A server described in any one of (1) to (5), wherein the second amount is calculated based on an amount calculated based on the first amount and the specified ratio, and the hedged amount. (7) The server according to (6), wherein the limit order and the stop order are each calculated based on the hedge rate when the hedged amount is hedged and the ratio. (8) The server according to (1) or (2), wherein the first amount is the amount of the first currency to be received on the due date or the amount of the first currency to be paid on the due date. (9) The server according to any one of (1) to (8), wherein the first currency and the second currency include a combination of at least one different currency among the U.S. dollar, euro, Japanese yen, British pound, Chinese yuan, Swiss franc, Australian dollar, Singapore dollar, Indian rupee, Indonesian ringgit, Thai baht, Philippine peso, South Korean won, and Vietnamese dong. (10) A server according to any one of (1) to (9), wherein each of the plurality of AI models is a different AI model generated by deep learning using a plurality of teacher data, at least some of which are common, and varying the individual weights of the plurality of teacher data. (11) A program for causing a computer to function as a server according to any one of (1) to (10).

[0080] [Other embodiments] The invention is not limited to the above-described embodiments, and various modifications and variations are possible within the spirit and scope of the invention. Therefore, in order to clarify the scope of the invention, the following claims are appended. The information processing device according to the present invention can also be realized by a computer program that causes one or more computers to function as the information processing device. The computer program can be provided / distributed by being recorded on a computer-readable recording medium or via a telecommunications line. [Explanation of symbols]

[0081] 10: System, 101: Server, 102: User terminal, 103: Financial services business system, 104: Network

Claims

1. A server that provides information about currency hedging to a user terminal, one or more processors; Memory and a program stored in the memory that, when executed by the one or more processors, Using a plurality of AI models to obtain, for each AI model, a forecast of future movements in an exchange rate indicating the value of a second currency relative to a first currency; accepting a designation of a first amount of the first currency from the user terminal; receiving input of at least one of a first desired exchange rate and a second desired exchange rate for the first amount of foreign exchange reservation, and an input of a maturity date of the foreign exchange reservation; calculating a limit price for the exchange rate for a forward foreign exchange contract of the first amount based on the received first desired rate, the first amount, and a hedged amount already hedged in relation to the maturity date; calculating a stop loss of the exchange rate for the forward exchange contract of the first amount based on the received second desired rate, the first amount, and the hedged amount; Calculating a first realization probability of the calculated limit price during the period up to the expiration date based on future trends in the exchange rate individually predicted by the plurality of AI models; Calculating a second realization probability of the calculated stop loss order during the period up to the expiration date based on future trends in the exchange rate individually predicted by the plurality of AI models; transmitting information on a second amount to be subject to currency hedging among the first amount to the user terminal together with information on the calculated limit order and the calculated first realization probability, and information on the calculated stop order and the calculated second realization probability; The server that runs the

2. The first realization probability is calculated based on a ratio of AI models among the plurality of AI models that predicted an event in which the exchange rate reaches the limit price within the period until the due date; 2. The server according to claim 1, wherein the second realization probability is calculated based on a proportion of the AI ​​models among the plurality of AI models that predicted an event in which the exchange rate reaches the stop price within the period until the due date.

3. Receiving designation of a first amount of the first currency from the user terminal includes: receiving, from the user terminal, a designation of a third amount of the first currency to be paid on the due date; accepting, from the user terminal, a designation of a fourth amount of the first currency to be received on the due date; Including, The server of claim 2 , wherein the difference between the third amount and the fourth amount is designated as the first amount.

4. If the sign of the first amount is positive, a second desired rate is specified as an amount by which the exchange rate has changed from the current exchange rate in a first direction that devalues ​​the second currency; 4. The server of claim 3, wherein if the sign of the first amount is negative, the second desired rate is specified as an amount of change in the exchange rate from the current exchange rate in a second direction in which the value of the second currency increases.

5. The program, when executed by the one or more processors, causes the server to:

5. The server according to claim 4, further configured to transmit to the user terminal, together with information on the second amount, a notification indicating that an exchange reservation at the current exchange rate is recommended, instead of information on the limit price and the first realization probability and information on the stop loss price and the second realization probability, when the sign of the first amount is positive and the amount indicated by the limit price is higher than the amount based on the current exchange rate, and when the sign of the first amount is negative and the amount indicated by the limit price is lower than the amount based on the current exchange rate.

6. When executed by the one or more processors, the program further causes the server to receive an input specifying a proportion of the first amount to be subject to the currency hedging; The server according to claim 5 , wherein the second amount is calculated based on an amount calculated based on the first amount and the specified ratio, and the hedged amount.

7. The server according to claim 6 , wherein the limit price and the stop price are each calculated further based on a hedge rate at which the hedged amount is hedged and the ratio.

8. The server according to claim 2 , wherein the first amount is an amount of the first currency to be received on the due date or an amount of the first currency to be paid on the due date.

9. 8. The server of claim 7, wherein the first currency and the second currency include at least any combination of different currencies selected from the group consisting of United States dollars, euros, Japanese yen, British pounds, Chinese yuan, Swiss francs, Australian dollars, Singapore dollars, Indian rupees, Indonesian ringgits, Thai baht, Philippine pesos, South Korean wons, and Vietnamese dongs.

10. The server according to claim 9, wherein each of the plurality of AI models is a different AI model generated by deep learning using a plurality of teacher data, at least some of which are common, and varying the individual weights of the plurality of teacher data.

11. A computer program for causing a computer to function as the server according to any one of claims 1 to 10.

Images