Information processor, information processing method, and program

Abstract

To provide an information processor that estimates a price of a transaction target in a market of electric power or the like with higher accuracy.SOLUTION: An information processor includes a processing unit. The processing unit calculates a similarity between a first explanatory variable group and each of a plurality of second explanatory variable groups. The first explanatory variable group includes a first target price which is a value at a first retroactive time that is traced back from a first time by a retroactive time, and first related information which is a value at the first time of related information including at least one of a related price of a transaction target and weather information. The second explanatory variable group corresponds to a plurality of second times prior to the first time, and includes a second target price and second related information at a second retroactive time that is traced back from the second time by the retroactive time. The processing unit extracts one or more second explanatory variable groups from the plurality of second explanatory variable groups, the second explanatory variable groups having a degree of similarity greater than other second explanatory variable groups, and estimates, by using a third target price at the second time corresponding to the extracted second explanatory variable group, a fourth target price at the first time.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] Embodiments of the present invention relate to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] With the liberalization of the electricity market, it has been required to accurately estimate (predict) the wholesale price (unit price) of electricity and the imbalance charge. For example, methods for estimating the price based on past performance close to the time period and weather conditions to be estimated, and methods for estimating the price from time-series performance up to the immediate past have been proposed.

[0003] Note that the wholesale price of electricity includes, for example, the price of electricity in the spot market (spot price) and the price of electricity in the forward market. Since the imbalance charge is determined after the actual supply and demand, its nature is different from the wholesale price.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0005] An object of the present invention is to provide an information processing apparatus, an information processing method, and a program capable of estimating the price of a transaction target in a market such as electricity with higher accuracy.

Means for Solving the Problems

[0006] The information processing apparatus according to the embodiment includes a processing unit. The processing unit calculates the similarity between the first explanatory variable group and each of the plurality of second explanatory variable groups. The first explanatory variable group includes a first target price which is a value at a first retroactive time retroactively traced by a retroactive time from a first time, and a first related information which is a value at the first time of related information including at least one of a related price of a transaction target and weather information. The second explanatory variable group corresponds to a plurality of second times before the first time, and includes a second target price at a second retroactive time retroactively traced by a retroactive time from the second time and second related information. The processing unit extracts one or more second explanatory variable groups having a similarity greater than that of other second explanatory variable groups from the plurality of second explanatory variable groups, and uses the third target price at the second time corresponding to the extracted second explanatory variable group to estimate the fourth target price at the first time.

Brief Description of Drawings

[0007]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Embodiments for Carrying Out the Invention

[0008] With reference to the accompanying drawings, a preferred embodiment of an information processing apparatus according to the present invention will be described in detail. Hereinafter, an example in which the market and the trading target in the market are the power supply and demand adjustment market and power will be described, but the market and the trading target are not limited thereto.

[0009] After the full liberalization of power retail from 2016, under the planned value simultaneous same quantity system, retailers and power generators create power demand plans and power generation plans for each 30-minute period divided into 48 periods per day, and conduct operations to match supply and demand until one hour before actual supply and demand (gate closure).

[0010] In actual supply and demand, the deviation from the plan is called imbalance. When an imbalance occurs, the Transmission System Operator (TSO) issues a command to the adjustment power such as power sources and adjusts to eliminate the imbalance. Those who cause the imbalance will conduct ex-post settlement with the power transmission and distribution operator as an imbalance charge for the power corresponding to the imbalance.

[0011] By estimating this imbalance charge in advance, it is expected to lead to an improvement in the profitability of trading strategy algorithms for formulating highly profitable trading plans in the spot market and the pre-time market, and the profitability of the operation plan algorithm for storage batteries.

[0012] Note that hereinafter, an example in which the price to be estimated (target price) is the imbalance charge will be mainly described. The target price is not limited to the imbalance charge, and may be, for example, the price of power in the pre-time market, or both the imbalance charge and the price of power in the pre-time market. Note that hereinafter, the target price may sometimes be simply referred to as the power price.

[0013] Since there were issues such as cases where the imbalance charges before 2021 did not reflect the value of the power for actual supply and demand for each area and each period, a new imbalance charge system has started since 2022.

[0014] The following embodiments are applicable to a new imbalance charge system, and for example, enable the estimation (prediction) of the imbalance charge on the day of the transaction. The imbalance charge is such that the performance of the previous period is publicly disclosed as a quick report. Also, weather forecasts from just before supply and demand can be used. Therefore, by using at least a part of this information, it becomes possible to estimate the electricity price with higher accuracy. As a result, it is possible to improve the user's profit and achieve a stable supply of electricity.

[0015] For example, the information processing apparatus according to the embodiment estimates the electricity price (target variable) at the time T_A (first time) to be estimated, using the actual performance of the electricity price at the retroactive time BT_A (first retroactive time) that is retroactively traced back from the time T_A by the retroactive time, and the related information as explanatory variables. For example, the information processing apparatus according to the embodiment extracts a plurality of explanatory variables (performance variable lists) similar to the said explanatory variables from past performance data, and estimates the electricity price at the time T_A using the plurality of extracted performance variable lists. When electricity is the subject of a transaction, the time may be specified by a time corresponding to a 30 - minute period (time every 30 minutes).

[0016] The related information includes, for example, at least one of weather information and the spot price at the time T_A. The spot price is an example of a related price that is a price of the transaction target related to the electricity price (target price). The spot price includes one or both of a system price representing an index price at which supply and demand across the country balance, and an area price representing the price agreed upon in each area divided across the country. Hereinafter, an example in which both weather information and the spot price are used as related information will be mainly described.

[0017] FIG. 1 is a block diagram showing an example of the configuration of the information processing apparatus 100 according to the embodiment. As shown in FIG. 1, the information processing apparatus 100 includes an acquisition unit 101, a determination unit 102, an extraction unit 103, a calculation unit 104, an estimation unit 105, an output control unit 106, and a storage unit 120.

[0018] The acquisition unit 101 acquires various types of information used in the information processing apparatus 100. For example, the acquisition unit 101 acquires power price performance data and spot price performance data from an external device (such as a device managed by a power generation company) that manages these data. Further, for example, the acquisition unit 101 acquires, as weather information, weather forecast values predicted and output by an organization such as the Japan Meteorological Agency using a local forecast model (LFM). The acquisition unit 101 stores the acquired various types of information in, for example, the storage unit 120.

[0019] Also, the acquisition unit 101 acquires various conditions (estimation conditions) used for estimation, which are input by a user or the like using an input interface (such as a keyboard or a mouse). The acquisition unit 101 may acquire the estimation conditions from an external device.

[0020] The estimation conditions include, for example, the following conditions. · The number K of the list of performance variables extracted by the extraction unit 103 (or the similarity threshold) · The coefficient (weight) of each element included in the explanatory variable, which is used when the calculation unit 104 calculates the similarity · The estimation method when estimating the power price

[0021] The power price performance data is data representing the past performance of power prices in transactions. For example, the power price performance data includes the transaction time and the power price. The spot price performance data is data representing the past performance of spot prices in transactions. For example, the spot price performance data includes the transaction time and the spot price.

[0022] The weather information includes, for example, the following information. · At least one of the weather forecast value and the weather observation value at each time in past transactions · At least one of the weather forecast value and the weather observation value at each time up to the time T_A to be estimated

[0023] Generally, even if the weather prediction value is the latest one, errors can still occur. Therefore, for example, the estimation accuracy can be improved by using the latest weather observation value as an explanatory variable (related information). The weather observation value can be any observation value obtained in any way. For example, the weather observation value observed at an AMeDAS site by the Japan Meteorological Agency and the weather observation value managed by the company itself can be used.

[0024] The determination unit 102 determines the retroactive time. For example, the determination unit 102 determines the retroactive time based on one or more power prices obtained at one or more times before time T_A. More specifically, the determination unit 102 determines the retroactive time by, for example, the following method. · Determine the time for retroactively going back to one or more times including the latest time among one or more times before time T_A as one or more retroactive times. · If the time for retroactively going back to the latest time is greater than the threshold value (threshold value of the retroactive time), determine the retroactive time as 24 hours.

[0025] The extraction unit 103 extracts an estimation variable list and an actual result variable list, which are used for estimating the power price at time T_A, from the data (power price actual result data, spot price actual result data, and weather information) acquired by the acquisition unit 101.

[0026] The estimation variable list is information (first explanatory variable group) including a plurality of explanatory variables used for estimating the power price (target variable) to be estimated at time T_A. The estimation variable list includes, for example, the following elements (explanatory variables). Note that after each element, the data from which each element is acquired is described. · Power price P_A at time BT_A (first target price): Acquisition source = Power price actual result data · Weather information at time T_A (first related information): Acquisition source = Weather information · Spot price at time T_A (first related information): Acquisition source = Spot price actual result data

[0027] The performance variable list is information (second explanatory variable group) including a plurality of explanatory variables corresponding to a time T_B (second time) before time T_A. The time T_B may be a plurality of times. In this case, the extraction unit 103 extracts a plurality of performance variable lists (a plurality of second explanatory variable groups).

[0028] The performance variable list includes, for example, the following elements (explanatory variables). · The power price P_B (second target price) at the retroactive time BT_B (second retroactive time) retroactively calculated by the retroactive time from time T_B: Acquisition source = power price performance data · The weather information (first related information) at time T_B: Acquisition source = weather information · The spot price (first related information) at time T_B: Acquisition source = spot price performance data · The power price P_C (third target price) at time T_B: Acquisition source = power price performance data

[0029] As elements of the explanatory variables of the estimated variable list and the performance variable list (elements corresponding to the related information), at least one of time and information indicating whether it is a holiday (holiday determination) may be further included. For the holiday determination, for example, 1 may be specified for holidays (including national holidays) and 0 for non-holiday cases. Also, when a plurality of retroactive times are used, the amount of change in data at the times corresponding to the plurality of retroactive times may be included as an element of the explanatory variable.

[0030] FIG. 2 is a diagram showing an example of explanatory variables. In the example of FIG. 2, the explanatory variables include the actual value of the imbalance charge, the system price, the area price, the predicted temperature value, the predicted solar radiation amount value, the sine of time, the cosine of time, and the holiday determination.

[0031] The actual value of the imbalance charge corresponds to the power price P_A and the power price P_B. The system price and the area price correspond to the spot price. The predicted temperature value and the predicted solar radiation amount value correspond to the weather information. The sine of time and the cosine of time are an example of a representation method of time. Time may be represented by a method other than the representation method using sine and cosine.

[0032] Return to the description of FIG. 1. The calculation unit 104 calculates the similarity between the estimated variable list and each of the plurality of performance variable lists. For example, the calculation unit 104 calculates the similarity based on the distance between each element (power price P_A, weather information, spot price) included in the estimated variable list and the corresponding element (power price P_B, weather information, spot price) included in the performance variable list. When time, holiday determination, change amount, etc. are included as related information, the calculation unit 104 may also include the distance between these elements in the calculation of the similarity.

[0033] The distance between elements may be calculated by any method. For example, it is the Minkowski distance. The similarity is calculated, for example, by the reciprocal of the distance. The calculation unit 104 may calculate the similarity based on the value obtained by multiplying the distance by the coefficient defined for each element. The coefficient of each element is specified, for example, as an estimation condition.

[0034] The extraction unit 103 further extracts one or more performance variable lists from the plurality of performance variable lists, where the calculated similarity is greater than that of other performance variable lists. For example, the extraction unit 103 extracts one or more performance variable lists based on the estimation conditions (the number K of performance lists or the similarity threshold). When the number K of performance lists is used, the extraction unit 103 extracts K performance variable lists in descending order of similarity. When the similarity threshold is used, the extraction unit 103 extracts the performance variable lists whose similarity is greater than the threshold.

[0035] The estimation unit 105 estimates the power price P_D (the fourth target price) at time T_A using the power price P_C at time T_B corresponding to the extracted performance variable list. For example, the estimation unit 105 estimates the simple average value or the weighted average value of the power price P_C as the power price P_D.

[0036] How to calculate the electricity price \(P_D\) is specified by, for example, the estimation method included in the estimation conditions. For example, the estimation method is specified as either a method of estimating the simple average value of the electricity price \(P_C\) as the electricity price \(P_D\) or a method of estimating the weighted average value of the electricity price \(P_C\) as the electricity price \(P_D\).

[0037] FIG. 3 is a diagram showing an example of the estimation conditions used in each of the above processes. In the example of FIG. 3, the estimation conditions include the number of variable lists, the estimation method, and the coefficients (weights) of each element. The number of variable lists corresponds to the number \(K\) of performance lists extracted by the extraction unit 103.

[0038] The processes by the extraction unit 103, the calculation unit 104, and the estimation unit 105 (hereinafter, the estimation process) can also be interpreted as processes using the \(K\)-nearest neighbor method. The \(K\)-nearest neighbor method is a technique for obtaining an estimated value by selecting the \(K\) points closest to the point to be estimated in the explanatory variable space and integrating (ensembling) the target variables corresponding to the selected \(K\) points. Details of the estimation process will be described later.

[0039] Returning to the description of FIG. 1. The output control unit 106 controls the output of various information used in the information processing apparatus 100. For example, the output control unit 106 outputs the result of the estimation process (estimation result) to a display device included in the information processing apparatus 100 and an external device, etc.

[0040] At least a part of the above-described respective units (acquisition unit 101, determination unit 102, extraction unit 103, calculation unit 104, estimation unit 105, and output control unit 106) may be realized by one or more processing units. The above-described respective units may be realized by, for example, one or a plurality of processors. For example, the above-described respective units may be realized by causing a processor such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit) to execute a program, that is, by software. The above-described respective units may be realized by a processor such as a dedicated IC (Integrated Circuit), that is, by hardware. The above-described respective units may be realized by using software and hardware in combination. When using a plurality of processors, each processor may realize one of the respective units, or may realize two or more of the respective units.

[0041] The storage unit 120 stores various information used in the information processing apparatus. For example, the storage unit 120 stores power price performance data 121, spot price performance data 122, weather information 123, an estimation variable list 124, an actual variable list 125, and an estimation result 126.

[0042] Note that the storage unit 120 can be configured by any generally used storage medium such as a flash memory, a memory card, a RAM (Random Access Memory), an HDD (Hard Disk Drive), and an optical disk.

[0043] Each data (power price performance data 121, spot price performance data 122, weather information 123, estimation variable list 124, actual variable list 125, estimation result 126) may be stored in physically different storage media, or may be stored in different storage areas of a physically identical storage medium.

[0044] The information processing apparatus 100 may be physically configured by one device or may be physically configured by a plurality of devices. For example, the information processing apparatus 100 may be constructed in a cloud environment. Also, each unit within the information processing apparatus 100 may be provided in a distributed manner across a plurality of devices.

[0045] Next, information processing by the information processing apparatus 100 of the embodiment will be described. FIG. 4 is a flowchart showing an example of information processing in the embodiment.

[0046] The acquisition unit 101 acquires power price performance data, spot price performance data, and weather information (step S101). Also, the acquisition unit 101 acquires estimation conditions (step S102). The acquired data is stored in the storage unit 120 as, for example, power price performance data 121, spot price performance data 122, and weather information 123.

[0047] The extraction unit 103 extracts an estimation variable list for the estimation target time T_A from the performance data (power price performance data, spot price performance data) (step S103). Also, the extraction unit 103 extracts a plurality of performance variable lists for a plurality of times T_B before the time T_A from the performance data (power price performance data, spot price performance data) (step S104). The extracted data is stored in the storage unit 120 as, for example, an estimation variable list 124 and a performance variable list 125.

[0048] The calculation unit 104 calculates the similarity between the estimation variable list and each of the plurality of performance variable lists (step S105). The extraction unit 103 extracts one or more performance variable lists with high similarity (step S106).

[0049] The estimation unit 105 estimates the power price P_D at the time T_A using the one or more extracted performance variable lists (step S107). The estimated power price P_D is stored in the storage unit 120 as, for example, an estimation result 126.

[0050] The output control unit 106 outputs the estimated electricity price P_D (step S108) and ends the process. The output control unit 106 may output the estimated electricity price in tabular form or in graph form.

[0051] FIG. 5 is a diagram showing an example of a display screen for displaying the estimation result. The display screen of FIG. 5 is an example of a screen that displays the estimated electricity price P_D in tabular form for a plurality of areas for a plurality of times T_A. Note that the electricity price for each area can be calculated by the average of the area prices (simple average value or weighted average value).

[0052] FIG. 6 is a diagram showing another example of a display screen for displaying the estimation result. The display screen of FIG. 6 is an example of a screen that displays the change in the electricity price P_D (estimated value) at a plurality of times for two areas (Hokkaido and Tohoku) in graph form.

[0053] Next, the details of the estimation process by the extraction unit 103, the calculation unit 104, and the estimation unit 105 will be further described.

[0054] As described above, the estimation process corresponds to a process using the K-nearest neighbor method. The performance data (electricity price performance data, spot price performance data) from which the extraction unit 103 extracts the list of performance variables corresponds to the learning data in the K-nearest neighbor method. The performance data (learning data) may be performance data obtained in a certain past period (sample period). That is, the acquisition unit 101 may acquire the electricity price performance data and the spot price performance data obtained in a certain past period (for example, the past three months). The sample period may not be a certain past period, but for example, the entire period in which the performance data was obtained. In such a case, the date and time when the performance data was obtained may also be an element (related information) of the explanatory variable, and the distance with respect to the date and time may also be considered when calculating the similarity.

[0055] The k-nearest neighbor method only memorizes the training data and does not require computational processing during learning. Therefore, it can be said that this method incurs low operational costs in price estimation related to the electricity market where trends are likely to change due to institutional changes or the like. Also, since it is a non-parametric method, high-nonlinearity estimation can be performed as long as there is training data.

[0056] In the k-nearest neighbor method, for example, the number of ensembles, the metrics of the ensemble, and the scaling factor that multiplies each element of the explanatory variables to adjust the distance in the feature space are hyperparameters. These hyperparameters correspond to the above estimation conditions as follows. Note that the hyperparameters (estimation conditions) may be optimized in advance using an optimization tool or the like. · Number of ensembles: The number K of the list of performance variables · Ensemble metrics: The estimation method when estimating the electricity price · Scaling factor: The coefficient (weight) of the element

[0057] Next, the details of the determination process of the retroactive time by the determination unit 102 will be described. The retroactive time can be determined based on, for example, the similarity between a plurality of frames. First, the autocorrelation coefficient will be described as an example of an index for confirming similarity.

[0058] The autocorrelation coefficient ρ is calculated by, for example, the following equation (1). X t represents the value of the variable at time t. μ and σ represent the average value and the standard deviation of the variable, respectively. τ represents the time difference (time lag) between the times of the two variables for which autocorrelation is calculated. The autocorrelation coefficient ρ can be interpreted as an index representing the similarity between two times (t and t - τ) in the same signal (variable).

Equation

[0059] FIG. 7 shows a graph in which the autocorrelation coefficient of the performance data of the imbalance charge is plotted with the time difference τ on the horizontal axis.

[0060] One of the characteristics of the autocorrelation coefficient of the imbalance charge is that it has an oscillation component with a 24-hour period. This oscillation component is due to the imbalance charge being affected by variables with a 24-hour period such as solar irradiance, temperature, and human activities.

[0061] As shown in FIG. 7, as the time difference τ increases from 0, the autocorrelation coefficient decreases. However, as it approaches the same frame of the previous day (the time when the time difference τ is 24 hours), there is a time when the autocorrelation coefficient starts to increase, and the autocorrelation coefficient reaches a maximum at the same frame of the previous day.

[0062] In the example of FIG. 7, the autocorrelation coefficient with respect to the same frame of the previous day is about 0.5. Also, for the performance data up to about 5 hours before the current day, the autocorrelation coefficient is maintained at this value (0.5) or higher.

[0063] Also, since the imbalance charge is published within 30 minutes after the actual supply and demand frame, even assuming that it takes 30 minutes to obtain the performance data, for example, the performance data of the frame 90 minutes before can be used for estimation. Therefore, for example, 90 minutes before can be set as the reference for the retroactive time (the shortest retroactive time).

[0064] From the above analysis, it is expected that the performance data of the imbalance charge on the current day is effective for estimating the electricity price from, for example, 90 minutes later to about 5 hours later.

[0065] The determination unit 102 may determine the retroactive time based on such analysis, for example. For example, the determination unit 102 determines the time for retroactive to each time from the latest time among the times when the performance data is obtained to the time when the autocorrelation coefficient becomes the same as that of the same frame of the previous day (in the example of FIG. 7, the time 5 hours before) as the retroactive time.

[0066] When the latest time is before the time at which the autocorrelation coefficient is about the same as that of the same frame on the previous day (the time five hours before in the example of FIG. 7), the determination unit 102 may determine the retroactive time to be 24 hours. This process corresponds to determining the retroactive time to be 24 hours when the time for retroactive to the latest time is greater than the threshold value of the retroactive time (five hours in the example of FIG. 7). The determination unit 102 may obtain the accuracy of the estimation process when the retroactive time is changed, and determine the retroactive time that provides the desired accuracy.

[0067] Next, the concept of the estimation process according to this embodiment will be described. FIG. 8 is a diagram for explaining the concept of the estimation process.

[0068] In FIG. 8, the time when the estimation process starts corresponds to the estimation start time, and the time T_A to be estimated corresponds to the time to be estimated. The sample period 601 corresponds to the period for obtaining the actual data from which the actual variable list 610 and the estimated variable list 620 are extracted. The time t within the sample period corresponds to the time T_B before the time T_A (the time to be estimated). The time t' = t - τ corresponds to the retroactive time BT_B.

[0069] The actual variable list 610 includes the following elements. · The electricity price 611 at the retroactive time BT_B (time t') (corresponding to the electricity price P_B) · The spot price 612 at the time T_B · The weather information 613 at the time T_B · The electricity price 614 at the time T_B (corresponding to the electricity price P_C)

[0070] The estimated variable list 620 includes the following elements. · The electricity price 621 at the retroactive time obtained by retroactively from the time T_A (the time to be estimated) by the retroactive time τ (corresponding to the electricity price P_A) · The spot price 622 at the time T_A · The weather information 623 at the time T_A

[0071] The calculation unit 104 calculates the similarity between the estimated variable list 620 and the actual variable list 610. Although one actual variable list 610 is shown in FIG. 8, a plurality of actual variable lists corresponding to a plurality of times t can be extracted from the sample period 601. The calculation unit 104 calculates the similarity between each of the plurality of actual variable lists and the estimated variable list 620.

[0072] The extraction unit 103 extracts K actual variable lists from the plurality of actual variable lists based on the similarity. FIG. 8 shows an example in which four (K = 4) actual variable lists are extracted.

[0073] The estimation unit 105 estimates the electricity price 631 (corresponding to the electricity price P_D) at time T_A using the electricity prices 614a to 614d included in the extracted actual variable lists.

[0074] As described above, the electricity price may be, for example, an imbalance charge, or may be the price of electricity in the time - ahead market. The price of electricity in the time - ahead market is not determined as one value for each time slot. Therefore, for example, the average value or the closing value of the price of each time slot is estimated as the electricity price of that time slot.

[0075] As shown in FIG. 8, in the present embodiment, the actual performance of the electricity price at a time retroactively traced back by the retroactive time from the estimation target time, and the related information (spot price, weather information) are used as explanatory variables to estimate the electricity price at the estimation target time. Thereby, the estimation of the electricity price can be executed with higher accuracy.

[0076] As described above, in the information processing apparatus of the embodiment, the price of a transaction target in a market such as electricity can be estimated with higher accuracy.

[0077] Next, the hardware configuration of the information processing apparatus of the embodiment will be described with reference to FIG. 9. FIG. 9 is an explanatory diagram showing an example of the hardware configuration of the information processing apparatus of the embodiment.

[0078] The information processing apparatus according to the embodiment includes a control device such as a CPU (Central Processing Unit) 51, a storage device such as a ROM (Read Only Memory) 52 and a RAM (Random Access Memory) 53, a communication I / F 54 that connects to a network and performs communication, and a bus 61 that connects each part.

[0079] The program executed by the information processing apparatus according to the embodiment is provided by being pre - incorporated in the ROM 52 or the like.

[0080] The program executed by the information processing apparatus according to the embodiment may be configured to be recorded on a computer - readable recording medium such as a CD - ROM (Compact Disk Read Only Memory), a flexible disk (FD), a CD - R (Compact Disk Recordable), a DVD (Digital Versatile Disk) in an installable or executable file format and provided as a computer program product.

[0081] Furthermore, the program executed by the information processing apparatus according to the embodiment may be configured to be stored on a computer connected to a network such as the Internet and downloaded via the network for providing. Also, the program executed by the information processing apparatus according to the embodiment may be configured to be provided or distributed via a network such as the Internet.

[0082] The program executed by the information processing apparatus according to the embodiment can cause a computer to function as each part of the above - described information processing apparatus. This computer can read a program from a computer - readable storage medium and execute it on the main storage device by the CPU 51.

[0083] A configuration example of the embodiment will be described below. (Configuration Example 1) A first explanatory variable group including a first target price, which is a value of a target price representing an estimated price for a transaction target at a first retroactive time retroactively traced back from a first time by a retroactive time, and a first related information, which is a value of related information including at least one of a related price, which is a price of the transaction target related to the target price, and weather information, at the first time. A plurality of second explanatory variable groups corresponding to a plurality of second times before the first time, each of the plurality of second explanatory variable groups including a second target price, which is a value of the target price at a second retroactive time retroactively traced back from the corresponding second time by the retroactive time, and a second related information, which is a value of the related information at the corresponding second time. Calculate the similarity of Extract one or more of the second explanatory variable groups among the plurality of second explanatory variable groups, for which the similarity is greater than that of the other second explanatory variable groups. Estimate a fourth target price, which is a value of the target price at the first time, using a third target price, which is a value of the target price at the second time corresponding to the extracted second explanatory variable group. Processing unit An information processing apparatus comprising (Configuration Example 2) The processing unit Determine the retroactive time based on one or more of the target prices obtained at one or more times before the first time. The information processing apparatus according to Configuration Example 1. (Configuration Example 3) The processing unit Determine, as one or more of the retroactive times, a time for retroactively tracing back to one or more times including the latest time among one or more times. The information processing apparatus according to Configuration Example 2. (Configuration Example 4) The processing unit When the time for retroactively tracing back to the latest time is greater than a threshold, determine the retroactive time to be 24 hours. The information processing apparatus according to Configuration Example 3. (Configuration Example 5) The related information includes the weather information. The weather information included in the first explanatory variable group includes the weather information that is a weather forecast value. The weather information included in each of the plurality of second explanatory variable groups includes the weather information that is at least one of a weather forecast value and a weather observation value. The information processing apparatus according to any one of Configuration Examples 1 to 4. (Configuration Example 6) The related information includes the weather information. The weather information included in the first explanatory variable group includes the weather information that is the latest weather observation value. The information processing apparatus according to any one of Configuration Examples 1 to 5. (Configuration Example 7) The transaction target is electric power. The target price is at least one of an imbalance charge and the price of electric power in the time - ahead market. The related price is the price of electric power in the spot market. The information processing apparatus according to any one of Configuration Examples 1 to 6. (Configuration Example 8) The related information further includes information indicating whether it is a holiday or not. The information processing apparatus according to any one of Configuration Examples 1 to 7. (Configuration Example 9) The processing unit calculates the similarity based on a value obtained by multiplying the distance between a plurality of elements included in the first explanatory variable group and the second explanatory variable group by coefficients determined for each of the plurality of elements. The information processing apparatus according to any one of Configuration Examples 1 to 8. (Configuration Example 10) The processing unit extracts one or more second explanatory variable groups based on at least one of the number of the second explanatory variable groups to be extracted and the similarity threshold value. The information processing apparatus according to any one of Configuration Examples 1 to 9. (Configuration Example 11) The processing unit estimates the simple average value or the weighted average value of the third target price as the fourth target price. The information processing apparatus according to any one of Configuration Examples 1 to 10. (Configuration Example 12) The processing unit a calculation unit that calculates the similarity; an extraction unit that extracts the second explanatory variable group; an estimation unit that estimates the fourth target price; and includes The information processing apparatus according to any one of Configuration Examples 1 to 11. (Configuration Example 13) An information processing method executed by an information processing apparatus, a first target price that is a value at a first retroactive time obtained by retroactively the target price representing the price to be estimated for a transaction target by a retroactive time from a first time, and a first related information that is a value at the first time of related information including at least one of a related price that is a price of the transaction target related to the target price and weather information, a first explanatory variable group including; a plurality of second explanatory variable groups corresponding to a plurality of second times before the first time, each of the plurality of second explanatory variable groups including a second target price that is a value of the target price at a second retroactive time obtained by retroactively the retroactive time from the corresponding second time, and a second related information that is a value of the related information at the corresponding second time, each of the plurality of second explanatory variable groups; a step of calculating the similarity of; a step of extracting one or more of the second explanatory variable groups in which the similarity is greater than the other second explanatory variable groups from the plurality of second explanatory variable groups; a step of estimating a fourth target price that is a value of the target price at the first time using a third target price that is a value of the target price at the second time corresponding to the extracted second explanatory variable group; An information processing method including. (Configuration Example 14) To the computer A first explanatory variable group including a first target price, which is a value of a target price representing an estimated price for a transaction target at a first retroactive time retroactively traced back from a first time by a retroactive time, and first related information, which is a value of related information including at least one of a related price, which is a price of the transaction target related to the target price, and weather information, at the first time. A plurality of second explanatory variable groups corresponding to a plurality of second times before the first time, each of the plurality of second explanatory variable groups including a second target price, which is a value of the target price at a second retroactive time retroactively traced back from the corresponding second time by the retroactive time, and second related information, which is a value of the related information at the corresponding second time. Calculating a similarity with respect to the above. Extracting one or more of the second explanatory variable groups among the plurality of second explanatory variable groups, in which the similarity is greater than that of the other second explanatory variable groups. Estimating a fourth target price, which is a value of the target price at the first time, using a third target price, which is a value of the target price at the second time corresponding to the extracted second explanatory variable group. A program for causing the above to be executed.

[0084] Although some embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalent scope thereof.

Explanation of Signs

[0085] 100 Information processing apparatus 101 Acquisition unit 102 Determination unit 103 Extraction unit 104 Calculation unit 105 Estimation unit 106 Output control unit 120 Memory unit 121 Electricity price performance data 122 Spot price performance data 123 Weather information 124 List of estimated variables 125 List of performance variables 126 Estimation result

Claims

1. A first explanatory variable group including a first target price which is a value of the target price representing a price to be estimated for a transaction target at a first retroactive time retroactively traced back from a first time by a retroactive time, and a first related information which is a value of the related information including at least one of a related price which is a price of the transaction target related to the target price and weather information at the first time; A plurality of second explanatory variable groups corresponding to a plurality of second times before the first time, wherein each of the plurality of second explanatory variable groups includes a second target price which is a value of the target price at a second retroactive time retroactively traced back from the corresponding second time by the retroactive time, and a second related information which is a value of the related information at the corresponding second time; calculate the similarity of; extract one or more of the second explanatory variable groups among the plurality of second explanatory variable groups, in which the similarity is greater than that of the other second explanatory variable groups; estimate a fourth target price which is a value of the target price at the first time, using a third target price which is a value of the target price at the second time corresponding to the extracted second explanatory variable group; a processing unit An information processing apparatus comprising.

2. The processing unit determines the retroactive time based on one or more target prices obtained at one or more times before the first time. The information processing apparatus according to claim 1.

3. The processing unit determines, as the one or more retroactive times, a time for retroactively tracing back to one or more times including the latest time among the one or more times. The information processing apparatus according to claim 2.

4. The processing unit when the time for retroactively tracing back to the latest time is greater than a threshold value, determines the retroactive time to 24 hours. The information processing apparatus according to claim 3.

5. The related information includes the weather information, the weather information included in the first explanatory variable group includes the weather information which is a weather forecast value, the weather information included in each of the plurality of second explanatory variable groups includes the weather information which is at least one of a weather forecast value and a weather observation value. The information processing apparatus according to claim 1.

6. The related information includes the weather information, the weather information included in the first explanatory variable group includes the weather information which is the latest weather observation value. The information processing apparatus according to claim 1.

7. The transaction target is electric power. The target price is at least one of an imbalance price and the price of electricity in the pre-time market, The related price is the price of electricity in the spot market, The information processing apparatus according to claim 1.

8. The related information further includes information indicating whether it is a holiday, The information processing apparatus according to claim 1.

9. The processing unit, calculates the similarity based on a value obtained by multiplying the distance between a plurality of elements included in the first explanatory variable group and the second explanatory variable group by coefficients defined for each of the plurality of elements, The information processing apparatus according to claim 1.

10. The processing unit, extracts one or more of the second explanatory variable groups based on at least one of the number of the second explanatory variable groups to be extracted and the threshold value of the similarity, The information processing apparatus according to claim 1.

11. The processing unit, estimates the simple average value or the weighted average value of the third target price as the fourth target price, The information processing apparatus according to claim 1.

12. The processing unit, includes a calculation unit that calculates the similarity, an extraction unit that extracts the second explanatory variable group, and an estimation unit that estimates the fourth target price, and is provided with, The information processing apparatus according to claim 1.

13. An information processing method executed by an information processing apparatus, a first target price that is a value at a first retroactive time obtained by retroactively the target price representing the price to be estimated for a transaction target by a retroactive time from a first time, and a first related information that is a value at the first time of related information including at least one of a related price that is the price of the transaction target related to the target price and weather information, a first explanatory variable group; a plurality of second explanatory variable groups corresponding to a plurality of second times before the first time, each of the plurality of second explanatory variable groups including a second target price that is a value of the target price at a second retroactive time obtained by retroactively the retroactive time from the corresponding second time, and a second related information that is a value of the related information at the corresponding second time, each of the plurality of second explanatory variable groups; a step of calculating the similarity with, a step of extracting one or more of the second explanatory variable groups among the plurality of second explanatory variable groups, the similarity of which is greater than that of the other second explanatory variable groups; a step of estimating a fourth target price that is a value of the target price at the first time using a third target price that is a value of the target price at the second time corresponding to the extracted second explanatory variable group, An information processing method including

14. to a computer, a first explanatory variable group including a first target price which is a value of the target price representing an estimated price for a transaction target at a first retroactive time retroactively traced back from a first time by a retroactive time, and a first related information which is a value of the related information including at least one of a related price which is a price of the transaction target related to the target price and weather information at the first time; a plurality of second explanatory variable groups corresponding to a plurality of second times before the first time, each of the plurality of second explanatory variable groups including a second target price which is a value of the target price at a second retroactive time retroactively traced back from the corresponding second time by the retroactive time, and a second related information which is a value of the related information at the corresponding second time; a step of calculating a similarity therebetween; a step of extracting one or more of the second explanatory variable groups among the plurality of second explanatory variable groups having a greater similarity than other second explanatory variable groups; a step of estimating a fourth target price which is a value of the target price at the first time by using a third target price which is a value of the target price at the second time corresponding to the extracted second explanatory variable group; a program for causing the above to be executed.

Images