Transaction behavior evaluation apparatus

Abstract

To provide a transaction behavior evaluation device capable of evaluating a transaction behavior not limited to a transaction actually performed in a market.SOLUTION: In a transaction action evaluation device 100, a strategic action model setting processing part 102 prepares strategic action data 151 for which a strategic action model being the model of the strategic action of a market participant is determined, and a peculiar action model setting processing part 103 prepares peculiar action data 152 for which a peculiar action model being the model of peculiar action being action different from the strategic action model is determined. A market simulation processing part 105 simulates a market according to the bidding action of a market participant determined from strategy action data 151 and peculiar action data 152, and action evaluation data 155 evaluates whether or not the market participant acts according to a strategy action model in the simulated market and the influence of the action taken by the market participant on the market.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a trading behavior evaluation device that evaluates the behavior of market participants in a trading market and monitors for fraud. [Background technology]

[0002] Technologies for evaluating and monitoring the trading behavior of market participants in trading markets include technologies for evaluating and monitoring the behavior of market participants in real time to detect fraudulent activities, technologies for monitoring market prices to detect abnormal changes, and technologies for analyzing the behavior of market participants based on trading performance data (e.g., Patent Documents 1 and 2).

[0003] The technology described in Patent Document 1 allows extraction conditions for multiple different transaction situations to be input, making it possible to easily and efficiently extract desired fraudulent transactions that match each extraction condition from a large amount of transaction data.

[0004] The technology in Patent Document 2 registers in advance the price movements of a product in the event of insider trading, and when the product after being traded on the market shows price movements that correspond to the registered price movements, it notifies the administrator that there is a risk that the transaction may constitute insider trading by displaying information about the transaction, such as the business that conducted the transaction. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-146093 [Patent Document 2] Japanese Patent Application Laid-Open No. 2007-272703 Summary of the Invention [Problem to be solved by the invention]

[0006] The technology disclosed in Patent Document 1 extracts fraudulent transactions from actual market transaction performance data or real-time transaction data. This technology only monitors and detects fraudulent transactions from the result data of actual transactions, and is unable to analyze or evaluate the impact of cases where the relevant bid was not made or where a bid that differed from the actual bid was made. As a result, the evaluation scope is limited to situations that actually occurred, and comparative analysis with situations that differed from the actual situation is not possible, and the effectiveness of this technology in evaluating the impact of the relevant bidding behavior on the market is limited.

[0007] The technology disclosed in Patent Document 2 determines whether a transaction constitutes insider trading after an actual transaction has been made. It also determines insider trading based on information such as the time when a change occurred in the product information of the transaction target, the time the transaction was made, and the magnitude and time of the price fluctuation. This technology only determines whether a bid that resulted in a successful transaction constitutes insider trading, and it is unable to compare and evaluate the results with actual transactions or evaluate the impact of trading behavior.

[0008] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a transaction behavior evaluation device that is capable of evaluating transaction behavior that is not limited to transactions that have actually been conducted. [Means for solving the problem]

[0009] The trading behavior evaluation device according to the present disclosure includes an information collection processing unit that acquires setting data used for market simulation, including market participant data indicating market participants and economic evaluation specification data that is data that determines the market saleable volume and market selling price of the market participants; a strategic behavior model setting processing unit that creates strategic behavior data based on a strategic behavior model set in accordance with a user's instruction; a peculiar behavior model setting processing unit that creates peculiar behavior data based on the peculiar behavior model set in accordance with a user's instruction as a model of peculiar behavior that is behavior different from the strategic behavior model; a behavior execution processing unit that determines bidding behavior in the market of the market participants based on the strategic behavior data and the peculiar behavior data; and a market simulation unit that simulates a market based on the bidding behavior of the market participants determined by the behavior execution processing unit, and performs execution processing for the bidding behavior of the market participants to create trading performance data. a processing unit, a behavior evaluation processing unit that generates behavior evaluation data by evaluating whether the market participants have acted in accordance with the strategic behavior model and the impact on the market of any idiosyncratic behavior taken by the market participants based on the profits of the market participants calculated from the economic feasibility judgment specification data and the trading performance data, and a result output processing unit that stores the behavior evaluation data in a data storage unit as evaluation result data, wherein the economic feasibility judgment specification data includes data on the demand for electricity to be supplied for each planning unit in a planning period, data on the type of power source, and information on parameters representing the specifications of the power source, and the market simulation processing unit further generates trading status data that is information regarding the trading status obtained in the market and includes one or more of the latest contract price in the market, past contract prices, and orders of other parties, and the behavior execution processing unit determines the bidding behavior of the market participants based on the trading status data. [Effects of the Invention]

[0010] The trading behavior evaluation device according to the present disclosure can evaluate trading behavior that differs from trading behavior performed in the market during market simulation, and evaluate the impact on the market, making it possible to perform a wide range of evaluations. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram of a transaction behavior evaluation system according to a first embodiment. [Figure 2] FIG. 1 is a diagram illustrating an example of a hardware configuration of a transaction behavior evaluation system. [Figure 3] 10 is a flowchart of a transaction behavior evaluation process according to the first embodiment. [Figure 4] FIG. 10 is a diagram illustrating an example of market participant data. [Figure 5] FIG. 10 is a diagram showing an example of information on power demand included in economic efficiency judgment specification data. [Figure 6] FIG. 10 is a diagram showing an example of information on a power source included in the economic efficiency judgment specification data. [Figure 7] FIG. 10 is a diagram showing an example of power supply parameters included in economic efficiency judgment specification data. [Figure 8] FIG. 10 is a diagram illustrating an example of strategic behavior data. [Figure 9] FIG. 10 is a diagram illustrating an example of peculiar behavior data. [Figure 10] 10 is a flowchart showing an example of behavior evaluation processing according to the first embodiment. [Figure 11] FIG. 10 is a block diagram of a transaction behavior evaluation system according to a second embodiment. [Figure 12] FIG. 10 is a diagram illustrating an example of behavioral performance data. [Figure 13] 10 is a flowchart of a transaction behavior evaluation process according to the second embodiment. [Figure 14] 10 is a flowchart of a behavior evaluation process according to the second embodiment. [Figure 15] FIG. 11 is a block diagram of a transaction behavior evaluation system according to a third embodiment. [Figure 16] FIG. 10 is a diagram illustrating an example of learning specification data. [Figure 17] 11 is a flowchart of a transaction behavior evaluation process according to the third embodiment. [Figure 18] 11 is a flowchart of a learning behavior model setting process according to the third embodiment. [Figure 19] 11 is a flowchart of a behavior evaluation process according to the third embodiment. [Figure 20] FIG. 10 is a block diagram of a transaction behavior evaluation system according to a fourth embodiment. [Figure 21] 13 is a flowchart of a transaction behavior evaluation process according to the fourth embodiment. [Figure 22] 13 is a flowchart of a behavior evaluation process according to the fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] <First Embodiment> An embodiment of a transaction behavior evaluation device according to the present disclosure will be described below. In the following description, the electricity market will be used as an example of a market.

[0013] 1 is a diagram showing the configuration of a transaction behavior evaluation system according to Embodiment 1. The transaction behavior evaluation system according to Embodiment 1 includes a transaction behavior evaluation device 100 and a data storage unit 200.

[0014] As shown in FIG. 1, the trading behavior evaluation device 100 includes an information collection processing unit 101, a strategic behavior model setting processing unit 102, a peculiar behavior model setting processing unit 103, a behavior execution processing unit 104, a market simulation processing unit 105, a behavior evaluation processing unit 106, and a result output processing unit 107.

[0015] The information collection processing unit 101 reads setting data 210 necessary for market simulation (transaction simulation) from the data storage unit 200. The setting data 210 is stored in advance in the data storage unit 200, and includes at least market participant data 211 and economic evaluation specification data 212.

[0016] The strategic behavior model setting processing unit 102 is a behavior model setting processing unit that creates a behavior model of market participants based on the market participant data 211 read by the information collection processing unit 101, and determines a model of strategic behavior (trading in the market, bidding price, bidding volume, etc.) for the market participants. Data on the strategic behavior model for market participants determined by the strategic behavior model setting processing unit 102 is output from the strategic behavior model setting processing unit 102 as strategic behavior data 151.

[0017] The idiosyncratic behavior model setting processing unit 103 determines idiosyncratic behaviors for market participants. Data on the idiosyncratic behaviors for market participants determined by the idiosyncratic behavior model setting processing unit 103 is output from the idiosyncratic behavior model setting processing unit 103 as idiosyncratic behavior data 152.

[0018] The behavior execution processing unit 104 determines bidding behavior of market participants in the market based on the strategic behavior data 151 and the peculiar behavior data 152. Furthermore, the market simulation processing unit 105 simulates the market based on the bidding behavior of the market participants determined by the behavior execution processing unit 104, performs contract processing for the bidding behavior of the market participants, and creates trading status data 154 and trading performance data 153.

[0019] The behavior evaluation processing unit 106 generates behavior evaluation data 155, which is data evaluating the behavior of market participants, based on the trading performance data 153.

[0020] The result output processing unit 107 stores the behavioral evaluation data 155 in the data storage unit 200 as evaluation result data 220. The result output processing unit 107 may also have a function of displaying the evaluation result data 220 stored in the data storage unit 200 on, for example, a display device (not shown) to present it to the user.

[0021] 2 is a diagram showing an example of the hardware configuration of a transaction behavior evaluation system. The transaction behavior evaluation system can be configured, for example, from a computer 10, and an input device 1, a network 2, an external storage device 3, and an output device 4 connected thereto.

[0022] The computer 10 may have a general configuration including a CPU (Central Processing Unit) 11, a main storage device 12 such as a memory, and an auxiliary storage device 13 such as a hard disk. The CPU 11 can, for example, deploy a program stored in the auxiliary storage device 13 onto the main storage device 12 and execute the program, and write the execution results to the main storage device 12 or the auxiliary storage device 13.

[0023] The input device 1 is a means for the user to input data to the computer 10, and is composed of, for example, a keyboard or a mouse. The external storage device 3 is a storage medium connected to the computer 10 via the network 2, and is composed of, for example, a hard disk, an optical disk, or a magneto-optical disk. Additional storage media (not shown) may be connected to the network 2. The output device 4 is a device that allows the computer 10 to present information to the user, and is, for example, a display device.

[0024] The information collection processing unit 101, strategic behavior model setting processing unit 102, and peculiar behavior model setting processing unit 103 of the trading behavior evaluation device 100 shown in Fig. 1 are used by a user to input data, and can be realized by the input device 1 shown in Fig. 2. Furthermore, the behavior execution processing unit 104, market simulation processing unit 105, behavior evaluation processing unit 106, and result output processing unit 107 are realized by the CPU 11 of the computer 10 executing a program.

[0025] 1 is for storing setting data 210 and evaluation result data 220, and can be realized by any one of main storage device 12, auxiliary storage device 13, network 2, and external storage device 3 in Fig. 2, or a combination of two or more of them. Also, output device 4 in Fig. 2 can be used by result output processing unit 107 shown in Fig. 1 to present evaluation result data 220 to a user.

[0026] 3 is a flowchart showing the processing of transaction behavior evaluation performed by the transaction behavior evaluation device 100 according to Embodiment 1. The method of transaction behavior evaluation performed by the transaction behavior evaluation device 100 will be described below with reference to FIG.

[0027] In the information collection process of step S101, the information collection processing unit 101 reads setting data 210 including market participant data 211 and economic evaluation specification data 212 from the data storage unit 200. The process of step S101 may be performed, for example, by the user inputting the setting data 210 using the input device 1, or by the user using the input device 1 to select one of a plurality of setting data 210 (for example, electronic files saved in CSV format) saved in advance in the auxiliary storage device 13 or the external storage device 3.

[0028] An example of market participant data 211 is shown in Figure 4. Market participant data 211 is data for distinguishing between market participants, and in this embodiment, as shown in Figure 4, it is data on the names of businesses participating in the market (hereinafter referred to as "business names"). The business names may be actual company names or may be temporary names. In addition to the business names, setting data 210 may also include information such as the region in which the business operates and the type of business.

[0029] The economic evaluation specification data 212 includes data on the demand for electricity to be supplied for each planned unit time during the planned period (hereinafter referred to as "demand data"), data on the type of power source (hereinafter referred to as "power source data"), and information on parameters that represent the specifications of the power source (hereinafter referred to as "power source parameters").

[0030] An example of demand data is shown in Figure 5. In the demand data, the planning period may be, for example, one week, and the planning unit time may be, for example, 30 minutes. If a business has no demand, the value of that consumer's electricity demand is set to 0.

[0031] An example of power source data is shown in Figure 6. Power source types include thermal power generators, hydroelectric power generators, and storage batteries. Other types of power sources may also include what is called demand response (or demand adjustment) (hereinafter referred to as "DR"), which increases or decreases demand to obtain supply capacity or adjustment capability, and power sources for which purchase contracts have been concluded with others. As shown in Figure 6, the power source data includes information on the business operators that own the power sources, in addition to the type of each power source. Note that information on businesses that do not own power sources does not appear in the power source data.

[0032] An example of power source parameters is shown in Figure 7. The power source parameters are linked to the power source shown in the power source data, and include information on the maximum output, minimum output, startup cost characteristics, and fuel cost characteristics of the power source.

[0033] The maximum output and minimum output of a power supply represent the range that the power supply can output, and are expressed by the following formula (1).

[0034]

number

[0035] In equation (1), t represents time, i represents the individual number of the power supply, Pmin represents the minimum output, Pmax represents the maximum output, U represents the power supply's activation / deactivation state (activated = 1, deactivated = 0), and P represents the power supply's output.

[0036] The startup cost characteristic is a characteristic used to calculate the cost required to start up a power source. The startup cost characteristic can be expressed as a fixed value or a polynomial. For example, it can be expressed as a quadratic expression of the length of downtime before startup, as in the following equation (2).

[0037]

number

[0038] In equation (2), SCost represents the power supply startup cost, H represents the continuous downtime before the power supply is started, and Sa, Sb, and Sc represent the coefficients of the power supply startup cost calculation formula. Alternatively, the startup cost characteristics can be expressed as a table showing the relationship between the power supply downtime and temperature and startup cost, and the power supply startup cost can be selected from that table. Specifically, for example, if Sa = 0.0, Sb = 0.0, and Sc = 1.0 x 10^6, the startup cost SCost is 1,000,000.

[0039] The fuel cost characteristic is a characteristic for determining the cost of fuel consumed when a power source requires fuel for power generation, and is expressed by a quadratic equation such as the following equation (3).

[0040]

number

[0041] In formula (3), FCost represents fuel cost, and Fa, Fb, and Fc represent coefficients in the fuel cost calculation formula. Specifically, for example, if the power source is operating with Fa = 0.002, Fb = 1.5, Fc = 3000, and P = 500, then FCost = 0.002 × 500 × 500 + 1.5 × 500 + 3000 = 4250. Also, if the power source is stopped with P = 0, then FCost = 0. Formula (3) calculates fuel cost directly from the output of the power source, but the method of calculating fuel cost is not limited to this. For example, fuel consumption may be calculated first and then multiplied by the unit price of fuel to calculate fuel cost.

[0042] 3, in the strategic behavior model setting process of step S102, the strategic behavior model setting processor 102 sets a strategic behavior model in accordance with a user's instruction and creates strategic behavior data 151. The process of step S102 is performed, for example, by the CPU 11 executing a program for setting a strategic behavior model that has been loaded into the main storage device 12, the CPU 11 reading and processing the setting data 210 loaded into the main storage device 24 in accordance with the program, and storing the processing results in the main storage device 12. At this time, one of a plurality of setting data 210 (for example, electronic files saved in CSV format) previously saved in the auxiliary storage device 13 or the external storage device 3 that the user specifies using the input device 1 may be loaded into the main storage device 12.

[0043] An example of a strategic behavior model is shown in Figure 8. For example, a market participant with a power source can generate electricity using the power source and take the action of supplying it to meet customer demand or selling it in the market.

[0044] Here, we will explain a simple example where operator A's strategy at time 1 is "Profit Assurance 1" in the strategic behavior model of Figure 8. Assuming there is one power source, the remaining output after subtracting the demand value from the power source's maximum output becomes the amount available for sale on the market, and the following equation (4) holds. At this time, it is assumed that the power source in question is operating.

[0045]

number

[0046] In equation (4), P sell_vol is the marketable amount, P max is the maximum output of the power source, and Demand is the demand. If there is no upper limit on the bid quantity, this P sell_vol It is possible to use this as the sales bid amount as it is.

[0047] Similarly, based on the cost of power generation by the power source, the price of electricity to be sold on the market is calculated using the following equation (5).

[0048]

number

[0049] In equation (5), P sell_tanka is the market selling price, P tanka is the cost of power generation, and α is the added profit. In equation (5), α is treated as a parameter to be added, but it may also be treated as a parameter to be multiplied by α, for example, as in the following equation (6).

[0050]

number

[0051] The marketable volume P obtained in this way sell_vol and market selling price P sell_tanka are the sales bid quantity and the sales bid price, respectively, and the combination of these is the bid content.

[0052] 3, in the peculiar behavior model setting process of step S103, the peculiar behavior model setting processing unit 103 sets an peculiar behavior model as shown in Fig. 9 in accordance with the user's instructions and creates peculiar behavior data 152. The peculiar behavior model is a general term for behaviors that are not considered to be normally performed by market participants, and is not limited to the behaviors shown in Fig. 9.

[0053] Figure 9 shows examples of idiosyncratic behavior models: "showing," "buying up," "low-pricing," "holding back," "overselling," and "excessive purchases." Showing up refers to, for example, making a large number of bids, cancellations, or corrections without the intention of trading at prices that are not executed in order to artificially fluctuate market prices or induce other parties to trade, thereby misleading others into believing that trading is active. Buying up refers to, for example, abusive buying and hoarding or making large purchase bids to inflate market prices, thereby misleading others into believing that market trading is active. Low-pricing refers to, for example, selling one's own electricity at a price lower than the normal bid price, such as bidding to sell at a price lower than the unit price, with the intent of intentionally fluctuating market prices or interfering with other parties' sales. Holding back refers to, for example, not selling one's own electricity to the market in order to artificially inflate market prices or interfering with other parties' purchases. Overselling refers to, for example, selling a large amount of one's own electricity to the market in order to artificially lower market prices or interfering with other parties' sales. Overbuying means buying more from the market than one needs, for example, with the intent of artificially inflating market prices or preventing others from purchasing.

[0054] The anomalous behavior model may include not only fraudulent transactions but also trading behavior that is uneconomical for businesses. Examples of uneconomical trading behavior include low-price bidding, where a sales bid is made lower than the unit cost of power generation by the power source. Selling on the market results in losses for the business, so this is not behavior that businesses would undertake without justification. One reason for being forced to sell at low prices is that when a power source's output is greater than demand and there is excess power, one strategy is to set the selling price of the power at a low market purchase price in order to balance supply and demand. Other examples of anomalous behavior include holding back on sales even when there is power available to sell, excessive selling behavior where a sales bid is made for more power than is available, and excessive purchasing where a purchase bid is made for more power than is necessary.

[0055] 3, in the behavior execution process of step S104, the behavior execution processing unit 104 determines the behavior of the market participants in the market based on the strategic behavior data 151 and the peculiar behavior data 152. For example, a market participant following the strategic behavior model determines the market saleable volume P sell_vol Market selling price P sell_tanka On the other hand, market participants who follow the idiosyncratic behavior model, for example, may bid at the market selling price P sell_tanka It is determined that the buyer will take action aimed at selling at a higher price than the buyer.

[0056] The behavior of market participants according to a behavioral model that acts based on publicly available market information is determined using trading status data 154 created by the market simulation processing unit 105. The trading status data 154 indicates various information about the trading status available in the market, such as the latest contract price in the market, past contract prices in the market, and market order information in which orders from other parties are posted. For example, when placing a buy bid in the market, a market participant can take action aiming to purchase at a low price by adopting a strategy of placing a buy bid at a price obtained by multiplying the latest contract price by 0.9.

[0057] In the market simulation processing of step S105, the market simulation processing unit 105 receives orders from each market participant determined by the action execution processing unit 104, performs contract calculations, and creates the above-mentioned trading status data 154 and trading performance data 153, which is information on the orders and contract results of each market participant. For example, in the case of continuous trading, orders are taken in chronological order, and if there is a bid whose buying and selling price matches, it is considered to be a contract. If there is no match, the order remains in the market and is subject to judgment as to whether or not subsequent orders can be contracted. The market simulation processing is not limited to contract calculations in continuous trading, and may also be contract calculations in single-price auctions or multi-price auctions, for example.

[0058] In the behavior evaluation process of step S106, the behavior evaluation processing unit 106 evaluates the behavior of the market participants based on the information on the orders and execution results of each market participant indicated in the trading performance data 153, and creates behavior evaluation data 155, which is data on the evaluation results. The process of step S106 is performed, for example, by the CPU 11 executing a program for evaluating the behavior of the market participants, which is loaded in the main storage device 12, and by the CPU 11 reading and processing the trading performance data 153 loaded in the main storage device 24 in accordance with the program, and storing the processing results in the main storage device 12. At this time, one of a plurality of data (e.g., electronic files saved in CSV format) previously stored in the auxiliary storage device 13 or the external storage device 3, which the user specifies using the input device 1, may be loaded into the main storage device 12.

[0059] 10 is a flowchart showing a specific example of the behavior evaluation process. In this embodiment, the behavior evaluation process is performed by the behavior evaluation processing unit 106 by executing the flow shown in FIG.

[0060] In the transaction performance data acquisition process of step S11, the behavior evaluation processing unit 106 acquires the transaction performance data 153. In the evaluation criterion setting process of step S12, the behavior evaluation processing unit 106 sets the evaluation criterion in accordance with the user's instructions. The evaluation criterion can be set, for example, based on the profit from the transaction.

[0061] In the transaction profit calculation process in step S13, the behavior evaluation processing unit 106 calculates the profit based on the evaluation criteria. If the profit is calculated from the simple difference between the power generation unit price and the contract price, the profit can be calculated using the following formula (7).

[0062]

number

[0063] In equation (7), SellReturns is the sales revenue, P cont_tanka is the execution price, P cont_volIn reality, sales revenue is adjusted by fees, taxes, etc.

[0064] In step S14, the behavior evaluation processing unit 106 determines whether the behavior model of the market participant is an anomalous behavior model. If it is determined to be an anomalous behavior model, the behavior evaluation processing unit 106 determines whether the market participant performed the set anomalous behavior in the anomalous behavior execution determination in step S15, and evaluates whether the anomalous behavior performed by the market participant caused a change in the market situation in the anomalous behavior impact evaluation in step S16. Whether the market situation has changed due to the anomalous behavior can be evaluated, for example, by simulating the trading market if the anomalous behavior was not performed and comparing the simulation results with the market situation.

[0065] In step S17, the behavior evaluation processing unit 106 determines whether the behavior model of the market participant is a strategic behavior model. If it is determined to be a strategic behavior model, the behavior evaluation processing unit 106 determines whether the market participant has performed the set strategic behavior in the strategic behavior implementation determination in step S18.

[0066] Finally, in the behavior evaluation data creation process in step S19, the behavior evaluation processing unit 106 creates behavior evaluation data 155 that indicates the evaluation results of the behaviors of the market participants evaluated in the above processes.

[0067] 3 , in the result output process of step S107, the result output processing unit 107 creates evaluation result data 220 to be stored in the data storage unit 200 based on the behavioral evaluation data 155, and outputs the data to the data storage unit 200. The process of step S107 is performed, for example, by having the CPU 11 execute a program for creating the evaluation result data 220 stored in the main storage unit 12, having the CPU 11 read and process the transaction performance data 153 stored in the main storage unit 24 in accordance with the program, and then converting the behavioral evaluation data 155 resulting from the processing into an electronic file in, for example, CSV format, and storing it in the main storage unit 12 or the auxiliary storage unit 13, which serves as the data storage unit 200, or in the external storage unit 3. The result output processing unit 107 may also use the output device 4 (for example, a display device) to present the behavioral evaluation data 155 stored in the data storage unit 200 to a user.

[0068] The transaction behavior evaluation device 100 of the first embodiment uses a strategic behavior model to simulate strategic market participant behavior in actual transactions. Furthermore, it can also use an idiosyncratic behavior model to simulate idiosyncratic behavior that rarely occurs in actual transactions (e.g., fraudulent bidding behavior or uneconomical behavior for a business operator), thereby generating market conditions in which idiosyncratic behavior is implemented. The transaction behavior evaluation device 100 can also generate behavior evaluation data 155 that evaluates the behavior of market participants based on transaction performance data 153 generated by such market simulation. This behavior evaluation data 155 can provide a behavior evaluation for when market participants engage in idiosyncratic behavior in a simulated market environment. The transaction behavior evaluation device 100 can, for example, detect fraudulent transactions or uneconomical behavior, and can arbitrarily generate situations that rarely occur using only actual transaction performance data, thereby enabling evaluation and monitoring of market participant behavior across a wide range of anticipated situations.

[0069] The transaction behavior assessment device 100 may be realized by hardware as shown in Fig. 2, or may be realized as software such as a program that embodies the functions of the transaction behavior assessment device 100 on a computer, or as a recording medium on which the software is recorded. The recording medium may be any type, such as a hard disk, an optical disk, or a magneto-optical disk.

[0070] <Embodiment 2> Fig. 11 is a block diagram of a transaction behavior evaluation system according to Embodiment 2. The configuration of the transaction behavior evaluation system in Fig. 11 is different from the configuration in Fig. 1 in that behavioral performance data 213 is added to the setting data 210 stored in the data storage unit 200, and the strategic behavior model setting processing unit 102 and strategic behavior data 151 are replaced with a reproduction behavior model setting processing unit 108 and reproduction behavior data 156, respectively. Since the other elements are basically the same as those in Fig. 1, explanations that overlap with those in Embodiment 1 will be omitted below.

[0071] The behavioral history data 213 is record data of past behaviors taken by market participants in the real market or simulated market, as shown in Fig. 12, for example. If the user of this system is a general market participant, it is difficult to grasp all of the behaviors of others, and therefore the behavioral history data 213 may be a record of the market participant's own behavior. On the other hand, if the user is a market administrator, it is preferable that the behavioral history data 213 include records of the behaviors of all market participants, so long as it is possible to grasp the behaviors of all market participants in the market.

[0072] The reproduced behavior model setting processor 108 is a behavior model setting processor that creates a behavior model of market participants based on the setting data 210, and creates reproduced behavior data 156 based on the behavioral performance data 213 in accordance with user instructions. The reproduced behavior model setting processor 108 is used by the user to input data, and can be realized, for example, by the input device 1 shown in FIG. 2.

[0073] In the second embodiment, the behavior execution processing unit 104 determines bidding behavior of market participants in the market based on the peculiar behavior data 152 and the reproduced behavior data 156.

[0074] Fig. 13 is a flowchart showing the transaction behavior evaluation process performed by the transaction behavior evaluation device 100 according to Embodiment 2. In the flowchart of Fig. 13, step S102 is replaced with step S108 in the flowchart of Fig. 3.

[0075] In the information collection process of step S101, the information collection processing unit 101 reads setting data 210 including market participant data 211, economic evaluation specification data 212, and behavioral performance data 213 from the data storage unit 200. The process of step S101 may be performed, for example, by the user inputting the setting data 210 using the input device 1, or by the user using the input device 1 to select one of a plurality of setting data 210 (for example, electronic files saved in CSV format) saved in advance in the auxiliary storage device 13 or the external storage device 3.

[0076] In the reproduction behavior model setting process of step S108, the reproduction behavior model setting processing unit 108 sets a reproduction behavior model in accordance with a user's instruction and creates reproduction behavior data 156. The process of step S108 is performed, for example, by the CPU 11 executing a program for setting a reproduction behavior model loaded in the main storage device 12, the CPU 11 reading and processing the setting data 210 loaded in the main storage device 24 in accordance with the program, and storing the processing results in the main storage device 12. At this time, one of a plurality of setting data 210 (for example, electronic files saved in CSV format) previously saved in the auxiliary storage device 13 or the external storage device 3 that the user specifies using the input device 1 may be loaded into the main storage device 12.

[0077] The replicative behavior model is set, for example, so that market participants will take bidding actions similar to those they have taken in the past. The replicative behavior is carried out, for example, by referring to biddings that have earned large profits in the past, and the market-saleable amount P sell_vol The entire amount is set to be sold as a bid.

[0078] The operation of the peculiar behavior model setting processor 103 in the peculiar behavior model setting process of step S103 may be the same as that of the first embodiment, or may set a setting that market participants will not take specific actions based on the behavior history data 213. For example, this setting assumes a case where market participants will not take actions suspected of being fraudulent, based on the behavior history data 213 that includes actions suspected of being fraudulent. The specific actions that are set not to be taken are not limited to actions suspected of being fraudulent, and may be, for example, actions with large bid volumes.

[0079] In the behavior execution process of step S104, the behavior execution processing unit 104 determines the behavior of the market participants in the market based on the reproduced behavior data 156 and the peculiar behavior data 152. For example, a market participant following a reproduced behavior model is determined to imitate the behavior of past market participants.

[0080] In the market simulation processing of step S105, the market simulation processing unit 105 receives orders from each market participant determined by the action execution processing unit 104, as in embodiment 1, performs contract calculations, and creates trading status data 154 and trading performance data 153.

[0081] In the behavior evaluation process of step S106, the behavior evaluation processing unit 106, as in the first embodiment, evaluates the behavior of the market participants based on the information on the orders and execution results of each market participant indicated in the trading performance data 153, and creates behavior evaluation data 155, which is data on the evaluation results.

[0082] 14 is a flowchart showing a specific example of the behavior evaluation process in Embodiment 2. In this embodiment, the behavior evaluation process is performed by the behavior evaluation processing unit 106 executing the flow shown in FIG.

[0083] Steps S11 to S15 in Figure 14 are the same as those in Figure 10. In the anomalous behavior impact evaluation in step S16, the behavior evaluation processing unit 106 evaluates how the presence of a market participant following an anomalous behavior model has affected and changed the trading results of market participants who behaved according to a reproduced behavior model. Alternatively, the trading results of market participants who behaved according to an anomalous behavior model with some behavior changes based on the behavior history data 213 may be compared with the behavior history data to evaluate how the trading results of the market participants have changed.

[0084] In step S31, the behavior evaluation processing unit 106 determines whether the behavior model of the market participant is a reproduction behavior model. If it is determined to be a reproduction behavior model, the behavior evaluation processing unit 106 determines whether the market participant has performed the set reproduction behavior in the reproduction behavior execution determination in step S32.

[0085] In the behavior evaluation data creation process in step S19, the behavior evaluation processing unit 106 creates behavior evaluation data 155 that indicates the evaluation results of the behaviors of the market participants evaluated in the above process.

[0086] 13, in the result output process of step S107, the result output processing unit 107 creates evaluation result data 220 to be stored in the data storage unit 200 based on the behavioral assessment data 155, as in the first embodiment, and outputs the data to the data storage unit 200. The result output processing unit 107 may also use the output device 4 (for example, a display device) to present the behavioral assessment data 155 stored in the data storage unit 200 to the user.

[0087] The trading behavior assessment device 100 of the second embodiment can use a reproduced behavior model to reproduce transactions performed in an actual trading market, and can use an idiosyncratic behavior model to generate a market situation in which idiosyncratic behavior is performed. Furthermore, based on the idiosyncratic behavior model, it can also generate a model (a model in which specific behavior is not performed) in which specific behaviors of market participants are excluded from the reproduced behavior model, thereby creating a situation in which specific behaviors are not performed. The trading behavior assessment device 100 can also generate behavior assessment data 155 that evaluates the behavior of market participants based on trading performance data 153 created by simulating such a market. This behavior assessment data 155 not only reproduces the market but also provides behavior assessments when any idiosyncratic behavior model is added. For example, it is possible to arbitrarily evaluate the behavior of market participants when fraudulent transactions or uneconomical behaviors are added, or when suspected such behaviors are excluded. Therefore, by comparing and evaluating the actual trading situation, it is possible to evaluate and monitor the market situation based on the presence or absence of idiosyncratic behaviors and the resulting behavior of market participants.

[0088] <Third Embodiment> Fig. 15 is a block diagram of a transaction behavior evaluation system according to Embodiment 3. The configuration of the transaction behavior evaluation system in Fig. 15 is different from the configuration in Fig. 1 in that learned specification data 214 is added to setting data 210 stored in the data storage unit 200, and in that the strategic behavior model setting processing unit 102 and strategic behavior data 151 are replaced with a learned behavior model setting processing unit 109 and learned behavior data 157, respectively. Since the other elements are basically the same as those in Fig. 1, descriptions overlapping with those in Embodiment 1 will be omitted below.

[0089] FIG. 16 shows an example of the learning specification data 214. The learning specification data 214 in FIG. 16 is the behavioral performance data 213 shown in FIG. 12 with evaluation information on the behavioral details added. For example, when a market participant uses its own behavioral record as the learning specification data 214, this evaluation information may include information such as a profit evaluation based on trading results, whether or not there was more economical trading behavior, and the ease of closing a deal. Even if the user of the trading behavior evaluation system is a market administrator, they cannot directly know the profit information of market participants, but they can add evaluation information such as the possibility of fraudulent trading in bidding behavior.

[0090] The learned behavior model setting processor 109 is a behavior model setting processor that creates a behavior model of market participants based on the setting data 210, and sets the learned behavior data 157 according to the user's instructions. The learned behavior model setting processor 109 is used by the user to input data, and can be realized by, for example, the input device 1 shown in FIG. 2.

[0091] In the third embodiment, the behavior execution processing unit 104 determines the behavior of the market participants in the market based on the peculiar behavior data 152 and the learned behavior data 157.

[0092] Fig. 17 is a flowchart showing the transaction behavior evaluation process performed by the transaction behavior evaluation device 100 according to Embodiment 3. In the flowchart of Fig. 17, step S102 in the flowchart of Fig. 3 is replaced with step S109.

[0093] In the information collection process of step S101, the information collection processing unit 101 reads setting data 210 including market participant data 211, economic evaluation specification data 212, and learning specification data 214 from the data storage unit 200. The process of step S101 may be performed, for example, by the user inputting the setting data 210 using the input device 1, or may be performed by the user using the input device 1 to select one of a plurality of setting data 210 (for example, electronic files saved in CSV format) saved in advance in the auxiliary storage device 13 or the external storage device 3.

[0094] In the learning behavior model setting process of step S109, the learning behavior model setting processor 109 sets a learning behavior model, which is an action taken to achieve a specific purpose from past market movements, in accordance with a user's instructions, and creates learning behavior data 157. The process of step S109 is performed, for example, by the CPU 11 executing a program for setting a learning behavior model that has been loaded into the main storage device 12, the CPU 11 reading and processing the setting data 210 loaded into the main storage device 24 in accordance with the program, and storing the processing results in the main storage device 12. At this time, one of a plurality of setting data 210 (for example, electronic files saved in CSV format) previously saved in the auxiliary storage device 13 or the external storage device 3 that the user specifies using the input device 1 may be loaded into the main storage device 12.

[0095] As an example of learning, the learned behavior model setting processor 109 uses, for example, machine learning to extract past movements in the real market or simulated market from the learning specification data 214 and reflects the learned behavior. However, the learning method is not limited to machine learning, and the learning content is not limited to learning from past movements in the real market or simulated market. For example, the learned behavior model setting processor 109 can also import trading status data obtained from the market in real time, learn using reinforcement learning, and update the learned behavior data 157 as needed.

[0096] Here, the learning behavior model setting process when the learning behavior model setting processor 109 uses reinforcement learning will be described with reference to the flowchart shown in FIG.

[0097] First, in step S51, the learning behavior model setting processor 109 checks whether the current process is the initial setting. If it is the initial setting, the learning behavior model setting processor 109 acquires the learning specification data 214 in a learning specification data acquisition process in step S52, and sets the reward function in a reward function setting process in step S53.

[0098] As a method for setting the reward function, for example, when a power company is learning to take actions that seek economic benefits, it is necessary to set the reward function P, which is the power generation cost multiplied by a sufficient profit α. sell_tanka One possible method is to take the action of bidding and, if the bid is concluded in the market, a positive reward will be given since economic benefits will be gained, but if the bid is not concluded, no economic benefits will be gained and a zero or negative reward will be given.

[0099] On the other hand, if the current process is the second or subsequent time, the learned behavior model setting processor 109 acquires the market trading situation data 154 in a trading situation data acquisition process in step S54.

[0100] Next, in the reward calculation process of step S55, the learning behavior model setting processor 109 calculates the reward for the learning specification data 214 or the transaction status data 154. In the learning behavior data creation process of step S56, the learning behavior model setting processor 109 creates learning behavior data 157 including a behavior model such as bid price, bid amount, and bid timing based on the reward calculated in step S55.

[0101] For example, a simple learning model for supervised learning may be one in which, based on the economic evaluation specification data 212 and the learning specification data 214, a bid that will provide economic benefits is made, learning is performed to determine the result of a successful transaction as the correct answer, and learning behavior data is set to determine the bidding price and bid volume that will lead to the correct answer. However, the learning behavior model is not limited to a model that acts with the objective of obtaining economic benefits. For example, a learning behavior model may be one that primarily aims to balance supply and demand, lowers the priority from the perspective of economic benefits, and prioritizes completing a transaction in the market.

[0102] Returning to FIG. 17, in the peculiar behavior model setting process of step S103, the peculiar behavior model setting processor 103 can also realize peculiar behavior by taking a behavior that is not selected by the learned behavior model. For example, in the case of supervised learning, peculiar behavior can be generated by taking an action that is not the correct answer. As a specific example, it is possible to create peculiar behavior data 152 that disrupts the market by making a selling bid at an unreasonably low price without securing sufficient profit, or by making a selling bid in a volume that exceeds the capacity of the power source possessed.

[0103] In the behavior execution process of step S104, the behavior execution processing unit 104 determines the market behavior of the learned behavior model or the idiosyncratic behavior model based on the learned behavior data 157 and the idiosyncratic behavior data 152. For example, market participants following the learned behavior model will take actions aimed at obtaining economic merit or actions that prioritize closing a transaction in the market, in accordance with the content of their learning.

[0104] In the market simulation processing of step S105, the market simulation processing unit 105 receives orders from each market participant determined by the action execution processing unit 104, as in embodiment 1, performs contract calculations, and creates trading status data 154 and trading performance data 153.

[0105] In the behavior evaluation process of step S106, the behavior evaluation processing unit 106, as in the first embodiment, evaluates the behavior of the market participants based on the information on the orders and execution results of each market participant indicated in the trading performance data 153, and creates behavior evaluation data 155, which is data on the evaluation results.

[0106] 19 is a flowchart showing a specific example of the behavior evaluation process in Embodiment 3. In this embodiment, the behavior evaluation process is performed by the behavior evaluation processing unit 106 executing the flow shown in FIG.

[0107] Steps S11 to S15 in Figure 19 are the same as those in Figure 10. In the peculiar behavior impact evaluation in step S16, the behavior evaluation processing unit 106 evaluates how the presence of a market participant following an idiosyncratic behavior model has affected and changed the trading results of market participants who have behaved in accordance with the learned behavior model. Alternatively, the trading results of market participants who have behaved in accordance with the learned behavior model, with some of their behavior modified based on the behavior history data 213, may be compared with the behavior history data to evaluate how the trading results of the market participants have changed.

[0108] In step S61, the behavior evaluation processing unit 106 determines whether the behavior model of the market participant is a learned behavior model. If it is determined to be a learned behavior model, the behavior evaluation processing unit 106 determines whether the market participant has performed the set learned behavior in the learned behavior implementation determination in step S62.

[0109] In the behavior evaluation data creation process in step S19, the behavior evaluation processing unit 106 creates behavior evaluation data 155 that indicates the evaluation results of the behaviors of the market participants evaluated in the above process.

[0110] 17, in the result output process of step S107, the result output processing unit 107 creates evaluation result data 220 to be stored in the data storage unit 200 based on the behavioral assessment data 155, as in the first embodiment, and outputs the data to the data storage unit 200. The result output processing unit 107 may also use the output device 4 (for example, a display device) to present the behavioral assessment data 155 stored in the data storage unit 200 to the user.

[0111] The transaction behavior evaluation device 100 of the third embodiment can simulate a virtual business operator by using a learned behavior model to learn behaviors based on the learning specification data 214 and the economic evaluation specification data 212. Furthermore, it can simulate idiosyncratic behaviors not selected by the learned behavior model using an idiosyncratic behavior model. It can then generate behavior evaluation data 155 that evaluates the behavior of market participants based on the trading performance data generated by the market simulation. This behavior evaluation data 155 can not only simulate a virtual business operator, but also obtain behavior evaluations when any idiosyncratic behavior model is added. For example, it can arbitrarily evaluate the behavior of market participants when a turbulent market or uneconomical behavior is added. Therefore, by comparing and evaluating the actual trading situation, it is possible to evaluate and monitor idiosyncratic situations, market situations depending on the presence or absence of idiosyncratic behavior, and the behavior of market participants resulting from such situations.

[0112] <Fourth Embodiment> Fig. 20 is a block diagram of a transaction behavior evaluation system according to embodiment 4. The configuration of the transaction behavior evaluation system in Fig. 20 is the same as that in Fig. 1, except that the strategic behavior model setting processing unit 102 and strategic behavior data 151 in the transaction behavior evaluation device 100 are replaced with a behavior instruction setting processing unit 110 and behavior instruction data 158, respectively. Since the other elements are basically the same as those in Fig. 1, descriptions overlapping with those in embodiment 1 will be omitted below.

[0113] The action instruction setting processor 110 sets action instruction data 158 that defines action instructions regarding the trading behavior of market participants in accordance with user instructions and based on the trading status data 154. The action instruction setting processor 110 is used by the user to input data, and can be realized by, for example, the input device 1 shown in FIG.

[0114] In the fourth embodiment, the behavior execution processing unit 104 determines the behavior of the market participants in the market based on the peculiar behavior data 152 and the behavior instruction data 158.

[0115] Fig. 21 is a flowchart showing the transaction behavior evaluation process performed by the transaction behavior evaluation device 100 according to Embodiment 4. The flowchart in Fig. 17 is the same as the flowchart in Fig. 3 except that step S102 is replaced with step S110.

[0116] In the information collection process of step S101, the information collection processing unit 101 reads the setting data 210 including the market participant data 211 and the economic evaluation specification data 212 from the data storage unit 200, as in the first embodiment.

[0117] In the action instruction setting process of step S110, the action instruction setting processor 110 sets action instructions in accordance with the user's instructions and creates action instruction data 158. The process of step S110 is performed, for example, by the CPU 11 executing a program for setting action instructions that has been loaded into the main storage device 12, the CPU 11 reading and processing the setting data 210 that has been loaded into the main storage device 24 in accordance with the program, and storing the processing results in the main storage device 12. At this time, one of a plurality of setting data 210 (for example, electronic files saved in CSV format) that has been saved in advance in the auxiliary storage device 13 or the external storage device 3, that the user has specified using the input device 1, may be loaded into the main storage device 12.

[0118] The user sets action instructions using the action instruction setting processing unit 110 based on the trading situation data 154. For example, if there is sufficient time before closing, a user wishing to purchase will want to make a low bid, so the bid price is set to the market price multiplied by 0.95 by referring to the trading situation data 154. The bid volume for the market is set to, for example, the smaller of the sell bid volume published in the market and the company's available purchase volume by referring to the trading situation data 154.

[0119] In the peculiar behavior model setting process of step S103, the peculiar behavior model setting processing unit 103 sets behaviors that differ from the market behaviors of market participants following the behavior instructions as a peculiar behavior model, and creates peculiar behavior data 152.

[0120] In the behavior execution process of step S104, the behavior execution processing unit 104 determines the market participants' behavior in the market according to the behavior instruction and the peculiar behavior model based on the behavior instruction data 158 and the peculiar behavior data 152.

[0121] In the market simulation processing of step S105, the market simulation processing unit 105, as in the first embodiment, receives the orders of each market participant determined by the action execution processing unit 104, performs contract calculations, and creates the above-mentioned trading status data 154 and trading performance data 153, which is information on the orders and contract results of each market participant.

[0122] In the behavior evaluation process of step S106, the behavior evaluation processing unit 106, as in the first embodiment, evaluates the behavior of the market participants based on the information on the orders and execution results of each market participant indicated in the trading performance data 153, and creates behavior evaluation data 155, which is data on the evaluation results.

[0123] 22 is a flowchart showing a specific example of the behavior evaluation process in Embodiment 4. In this embodiment, the behavior evaluation process is performed by the behavior evaluation processing unit 106 executing the flow shown in FIG.

[0124] Steps S11 to S16 in Figure 22 are the same as those in Figure 10. In step S81, the behavior evaluation processing unit 106 determines whether or not there is a behavior instruction from the user. If it is determined that there is a behavior instruction, the behavior evaluation processing unit 106 determines whether or not the market participant acted in accordance with the behavior instruction in the behavior instruction implementation determination in step S82.

[0125] In the behavior evaluation data creation process in step S19, the behavior evaluation processing unit 106 creates behavior evaluation data 155 that indicates the evaluation results of the behaviors of the market participants evaluated in the above process.

[0126] 21, in the result output process of step S107, the result output processing unit 107 creates evaluation result data 220 to be stored in the data storage unit 200 based on the behavioral assessment data 155, as in the first embodiment, and outputs the data to the data storage unit 200. The result output processing unit 107 may also use the output device 4 (for example, a display device) to present the behavioral assessment data 155 stored in the data storage unit 200 to the user.

[0127] The transaction behavior evaluation device 100 of the fourth embodiment combines a behavior model that follows user behavioral instructions with an idiosyncratic behavior model, thereby creating situations in which behavior not performed in actual market transactions is performed using the idiosyncratic behavior model, and encouraging users to take actions based on judgments different from those in the actual market. Furthermore, it is possible to create behavior evaluation data that evaluates the behavior of market participants based on trading performance data created by market simulation. This behavior evaluation data allows users to consider actions they would take in situations that they would not experience in the actual market, and to obtain an evaluation of the results of their behavior. It is possible to arbitrarily evaluate situations that rarely occur in the actual market, such as market price increases, decreases, sudden changes, and insufficient bid volumes, and to comparatively evaluate a variety of trading situations.

[0128] In the fourth embodiment, the first embodiment is used as a reference and differences from the first embodiment are described. The fourth embodiment is intended for the user to create the behavior instruction data 158 at his / her discretion on the basis of the transaction situation data 154, and within the scope of this intention, the fourth embodiment may be combined with the peculiar behavior model that also uses the behavior record data 213 described in the second embodiment, or with the peculiar behavior model that also uses the learning specification data 214 described in the third embodiment, or with both of them.

[0129] In this way, it is possible to freely combine the respective embodiments, and to modify or omit the respective embodiments as appropriate. [Explanation of symbols]

[0130] 100 trading behavior evaluation device, 101 information collection processing unit, 102 strategic behavior model setting processing unit, 103 specific behavior model setting processing unit, 104 behavior implementation processing unit, 105 market simulation processing unit, 106 behavior evaluation processing unit, 107 result output processing unit, 108 reproduction behavior model setting processing unit, 109 learning behavior model setting processing unit, 110 behavior instruction setting processing unit, 151 strategic behavior data, 152 specific behavior data, 153 trading performance data, 154 trading status data, 155 behavior evaluation data, 156 reproduction behavior data, 157 learning behavior data, 158 behavior instruction data, 200 data storage unit, 210 setting data, 211 market participant data, 212 economic efficiency judgment specification data, 213 behavior performance data, 214 learning specification data, 220 evaluation result data, 1 input device, 2 network, 3 external storage device, 4 output device, 10 Computer, 11 CPU, 12 main memory, 13 auxiliary memory.

Claims

1. an information collection processing unit that acquires setting data used for market simulation, including market participant data indicating market participants and economic evaluation specification data that is data that determines the market saleable volume and market sale price of the market participants; a strategic behavior model setting processing unit that creates strategic behavior data based on a strategic behavior model set in accordance with a user's instruction; an idiosyncratic behavior model setting processing unit that creates idiosyncratic behavior data based on an idiosyncratic behavior model that is set in accordance with a user's instruction as a model of idiosyncratic behavior that is different from the strategic behavior model; a behavior execution processing unit that determines bidding behavior of the market participants in the market based on the strategic behavior data and the peculiar behavior data; a market simulation processing unit that simulates a market based on the bidding behavior of the market participants in the market determined by the behavior execution processing unit, and performs contract processing for the bidding behavior of the market participants to create trading performance data; a behavior evaluation processing unit that generates behavior evaluation data by determining whether the market participants have acted in accordance with the strategic behavior model and evaluating the impact of idiosyncratic behaviors taken by the market participants on the market, based on the profits of the market participants calculated from the economic evaluation specification data and the trading performance data; a result output processing unit that stores the behavior evaluation data as evaluation result data in a data storage unit; Equipped with The economic efficiency judgment specification data includes power demand data to be supplied for each planning unit during the planning period, data on the type of power source, and information on parameters representing the specifications of the power source, The market simulation processing unit further creates trading status data, which is information on trading status obtained in the market, including one or more of the latest contract price in the market, past contract prices, and orders of other parties; The behavior execution processing unit determines bidding behavior of the market participants in the market based on the trading status data. Trading behavior evaluation device.

2. an information collection processing unit that acquires setting data used for market simulation, including market participant data indicating market participants, economic evaluation specification data that determines the market saleable volume and market sale price of the market participants, and behavioral performance data of the market participants; a reproduction behavior model setting processing unit that creates reproduction behavior data that defines a reproduction behavior model, which is a model of reproduction behavior that is behavior that the market participant has previously performed in the market, based on the setting data; and a peculiar behavior model setting processing unit that creates peculiar behavior data that defines a peculiar behavior model that is a model of peculiar behavior that is different from the reproduced behavior model; a behavior execution processing unit that determines bidding behavior of the market participants in the market based on the reproduced behavior data and the peculiar behavior data; a market simulation processing unit that simulates a market based on the bidding behavior of the market participants in the market determined by the behavior execution processing unit, and performs contract processing for the bidding behavior of the market participants to create trading performance data; a behavior evaluation processing unit that generates behavior evaluation data by determining whether the market participants have acted in accordance with the reproduction behavior model and evaluating the impact of idiosyncratic behaviors taken by the market participants on the market, based on the profits of the market participants calculated from the economic efficiency judgment specification data and the trading performance data; a result output processing unit that stores the behavior evaluation data as evaluation result data in a data storage unit; Equipped with The economic evaluation specification data includes data on the demand for electricity to be supplied for each planning unit during the planning period, data on the type of power source, and information on parameters representing the specifications of the power source. Trading behavior evaluation device.

3. an information collection processing unit that acquires setting data used for market simulation, including market participant data indicating market participants, economic evaluation specification data that is data that determines the market saleable volume and market sale price of the market participants, and learning specification data that indicates an evaluation of the past actions of the market participants; a learning behavior model setting processing unit that creates learning behavior data that defines a learning behavior model, which is a model of learning behavior that is behavior taken to achieve a specific purpose from past market movements, based on the setting data; and a peculiar behavior model setting processing unit that creates peculiar behavior data that defines a peculiar behavior model that is a model of peculiar behavior that is different from the learned behavior model; a behavior execution processing unit that determines bidding behavior of the market participants in the market based on the learned behavior data and the peculiar behavior data; a market simulation processing unit that simulates a market based on the bidding behavior of the market participants in the market determined by the behavior execution processing unit, and performs contract processing for the bidding behavior of the market participants to create trading performance data; a behavior evaluation processing unit that generates behavior evaluation data by determining whether the market participants have acted in accordance with the learned behavior model and evaluating the impact of idiosyncratic behaviors taken by the market participants on the market, based on the profits of the market participants calculated from the economic evaluation specification data and the trading performance data; a result output processing unit that stores the behavior evaluation data as evaluation result data in a data storage unit; Equipped with The economic evaluation specification data includes data on the demand for electricity to be supplied for each planning unit during the planning period, data on the type of power source, and information on parameters representing the specifications of the power source. Trading behavior evaluation device.

4. an information collection processing unit that acquires setting data used for market simulation, including market participant data indicating market participants and economic evaluation specification data that is data that determines the market saleable volume and market sale price of the market participants; an action instruction setting processing unit that creates action instruction data that defines action instructions regarding trading actions of market participants set by a user; a peculiar behavior model setting processing unit that creates peculiar behavior data that defines a peculiar behavior model that is a model of peculiar behavior that is different from the behavior in accordance with the behavior instruction; a behavior execution processing unit that determines bidding behavior of the market participants in the market based on the behavior instruction data and the peculiar behavior data; a market simulation processing unit that simulates a market based on the bidding behavior of the market participants in the market determined by the behavior execution processing unit, and performs contract processing for the bidding behavior of the market participants to create trading performance data; a behavior evaluation processing unit that generates behavior evaluation data by determining whether the market participants have acted in accordance with the behavioral instructions and evaluating the impact of peculiar behaviors taken by the market participants on the market, based on the profits of the market participants calculated from the economic evaluation specification data and the trading performance data; and a result output processing unit that stores the behavior evaluation data as evaluation result data in a data storage unit; Equipped with The economic efficiency judgment specification data includes power demand data to be supplied for each planning unit during the planning period, data on the type of power source, and information on parameters representing the specifications of the power source, the market simulation processing unit further creates trading status data, which is information on trading status obtained in the market and includes one or more of the latest contract price in the market, past contract prices, and orders of other parties, and the user creates the action instruction data based on the trading status data; Trading behavior evaluation device.

5. The setting data further includes at least one of behavioral performance data of the market participants and learning specification data indicating evaluations of past behaviors of the market participants. The transaction behavior evaluation device according to claim 4.

6. The market simulation processing unit further creates trading status data, which is information on trading status obtained in the market, including one or more of the latest contract price in the market, past contract prices, and orders of other parties; The behavior execution processing unit determines bidding behavior of the market participants in the market based on the trading status data.

4. The transaction behavior evaluation device according to claim 2 or 3.

7. the result output processing unit further displays the evaluation result data on a display device. The transaction behavior evaluation device according to any one of claims 1 to 6.

Images