JP2021124802A - Automatic trade algorithm proposal program - Google Patents

Abstract

To propose automatic trade algorithms making profits.SOLUTION: An automatic trade algorithm proposal program for proposing automatic trade algorithms automatically performing stock trading, causes a computer to execute: a relation level obtainment step for obtaining a relation level, in a three or more steps, between each referential algorithm information relating to each automatic trade algorithm automatically performing the stock trading, and trade results having performed the automatic trade based on each referential algorithm information from an obtained stock chart in advance; a trade result extraction step for, from among the trade results relating to the relation levels obtained in the relation level obtainment step, prioritizing and extracting further excellent trade results; and a display step for prioritizing and displaying the referential automatic trade algorithms whose relation levels to the trade results extracted by the trade result extraction step are higher.SELECTED DRAWING: Figure 3

Description

The present invention relates to an automatic trading algorithm proposal program that proposes an automatic trading algorithm that automatically trades stocks.

In recent years, applications that provide various advice when conducting stock trading and systems that automate stock trading, that is, perform automatic trading, have been used. When using such an app or system, it is possible to increase the winning rate and increase profits by receiving advice based on a huge amount of data analysis. However, even if the enormous amount of data can be acquired, it takes considerable effort to analyze this and compile an output solution that can give accurate advice to the user. In addition, the current situation is that a system capable of automatically performing these operations has not been proposed in the past.

In recent years, an automatic trading algorithm that automatically trades stocks has also been proposed. There are many types of automatic trading algorithms, some of which make more profits on a case-by-case basis, while others make more profits and losses. For this reason, a tool that helps the user in selecting these automatic trading algorithms or in the operator side proposing the most suitable automatic trading algorithm to the user has been conventionally desired.

Therefore, the present invention has been devised in view of the above-mentioned problems, and an object thereof is in an automatic trading algorithm proposal program for proposing an automatic trading algorithm for automatically performing stock trading. The purpose is to provide an automatic trading algorithm proposal program that can provide useful information when selecting an automatic trading algorithm in the above, or when the operator side proposes the most suitable automatic trading algorithm to the user. ..

In order to solve the above-mentioned problems, in the automatic trading algorithm proposal program that proposes an automatic trading algorithm that automatically performs stock trading, each reference algorithm information regarding each automatic trading algorithm that automatically performs stock trading and acquired. It is related to the degree of association acquisition step that acquires in advance three or more levels of association with the transaction results that have been automatically traded based on each of the above reference algorithms from the stock price chart, and the degree of association acquired in the above association degree acquisition step. Among the above transaction results, a display that prioritizes and displays a transaction result extraction step that preferentially extracts a better transaction result and an automatic trading algorithm that has a higher degree of association with the transaction result extracted by the transaction result extraction step. It is characterized by having a computer perform steps.

Even if you do not have any special skills or experience, it is possible to propose an automatic trading algorithm that will generate more profits.

It is a block diagram which shows the whole structure of the system to which this invention is applied. It is a figure which shows the specific configuration example of a search device. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention. It is a figure for demonstrating the operation of this invention.

Hereinafter, the automatic trading algorithm proposal program to which the present invention is applied will be described in detail with reference to the drawings.

The first embodiment FIG . 1 is a block diagram showing an overall configuration of an automatic trade algorithm proposal system 1 in which an automatic trade algorithm proposal program to which the present invention is applied is implemented. The automatic trade algorithm proposal system 1 includes an information acquisition unit 9, a search device 2 connected to the information acquisition unit 9, and a database 3 connected to the search device 2.

The information acquisition unit 9 is a device for a person using this system to input various commands and information, and specifically, is composed of a keyboard, buttons, a touch panel, a mouse, a switch, and the like. The information acquisition unit 9 is not limited to a device for inputting text information, and may be configured by a device such as a microphone that can detect voice and convert it into text information. Further, the information acquisition unit 9 may be configured as an image pickup device capable of capturing an image of a camera or the like. The information acquisition unit 9 may be configured by a scanner having a function of recognizing a character string from a paper-based document. Further, the information acquisition unit 9 may be integrated with the search device 2 described later. The information acquisition unit 9 outputs the detected information to the search device 2. Further, the information acquisition unit 9 may be configured by means for specifying the position information by scanning the map information. The information acquisition unit 9 also includes a temperature sensor, a humidity sensor, a flow rate sensor, and other sensors capable of identifying substances and physical properties. The information acquisition unit 9 may be configured by means for automatically fetching character strings and data posted on a site on the Internet.

The database 3 stores various information necessary for proposing an automatic trading algorithm. Information necessary for proposing an automatic trading algorithm includes reference market information regarding past market conditions, reference event information that reflects events that occurred during the detection period of past market conditions, and external information regarding the detection time of past market conditions. External environmental information for reference that reflects the environment, household information for reference that reflects statistical data about households at the time of detection of past market conditions, and reference that reflects statistical data about real estate at the time of detection of past market conditions Real estate information, reference expert opinion information that reflects the opinions of experts announced at the time of past market condition detection, reference natural environment information that reflects information on the natural environment at the time of past market condition detection, A data set with the increase / decrease data of the stock price of each stock in the past market conditions is stored.

That is, in addition to such reference market condition information, the database 3 contains reference event information, reference external environment information, reference household information, reference real estate information, reference expert opinion information, and reference natural environment information. Any one or more of the above and the increase / decrease data of the stock price of each stock in the past market conditions are stored in association with each other.

The search device 2 is composed of, for example, an electronic device such as a personal computer (PC), but is embodied in any other electronic device such as a mobile phone, a smartphone, a tablet terminal, a wearable terminal, etc., in addition to the PC. It may be converted. The user can obtain a search solution by the search device 2.

FIG. 2 shows a specific configuration example of the search device 2. The search device 2 performs wired communication or wireless communication with a control unit 24 for controlling the entire search device 2 and an operation unit 25 for inputting various control commands via operation buttons, a keyboard, or the like. A communication unit 26 for the purpose, an estimation unit 27 for making various judgments, and a storage unit 28 for storing a program for performing a search to be executed represented by a hard disk or the like are connected to the internal bus 21, respectively. .. Further, a display unit 23 as a monitor that actually displays information is connected to the internal bus 21.

The control unit 24 is a so-called central control unit for controlling each component mounted in the search device 2 by transmitting a control signal via the internal bus 21. Further, the control unit 24 transmits various control commands via the internal bus 21 in response to the operation via the operation unit 25.

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

The estimation unit 27 estimates the search solution. The estimation unit 27 reads out various information stored in the storage unit 28 and various information stored in the database 3 as necessary information when executing the estimation operation. The estimation unit 27 may be controlled by artificial intelligence. This artificial intelligence may be based on any well-known artificial intelligence technique.

The display unit 23 is composed of a graphic controller that creates a display image based on the control by the control unit 24. The display unit 23 is realized by, for example, a liquid crystal display (LCD) or the like.

When the storage unit 28 is composed of a hard disk, predetermined information is written to each address based on the control by the control unit 24, and is read out as needed. Further, the storage unit 28 stores a program for executing the present invention. This program is read and executed by the control unit 24.

The operation in the automatic trading algorithm proposal system 1 having the above-described configuration will be described.

In the automatic trade algorithm proposal system 1, for example, as shown in FIG. 3, it is premised that the reference algorithm information and the transaction result are set and acquired in three or more stages in advance. The reference algorithm information is information on an algorithm for performing automatic trading of stock prices, crude oil, futures, precious metals, bitcoins, exchanges, etc. of each stock. Such an algorithm consists of trading rules. This trading rule also includes stock analysis, timing of trading, timing of settlement, and the like.

This algorithm includes those used in so-called algorithm trading in which a computer system automatically determines the timing and quantity of stock trading orders and repeats orders according to stock prices and trading volumes. For example, a buy / sell order may be dispersed so that the stock price does not fluctuate due to a transaction, or a buy order may be automatically placed at a timing when the stock price is judged to be cheap. Further, this algorithm may be used for so-called system trading. In system trading, trading is performed mechanically according to preset trading rules such as past market price movements. An algorithm that automatically places buy and sell orders based on the maximum rate of decline (drawdown) calculated from past stock price movements and the profit factor (PF), which indicates how many times the total profit is the total loss. It may be installed.

As an example of this algorithm, a stock whose "closing price" and "15-day moving average" have a deviation rate of more than -10% is sold or bought. As the closing condition, for example, an algorithm such as profit taking at the quoted price + 5%, loss cut at the quoted price -5%, and closing after 10 days of holding is shown.

The reference algorithm information may be such an algorithm itself or information for specifying this algorithm. As a method for specifying this algorithm, reference algorithm information may be constructed by the name or number when a name is assigned to each algorithm.

The transaction result is information on the result of trading stocks, exchanges, and futures based on the above-mentioned reference algorithm information for a certain predetermined period (for example, one month). This transaction result may be the winning rate in the predetermined period or the transaction profit / loss in the predetermined period. The transaction results shown in FIG. 3 are shown as + tens of thousands of yen and -tens of thousands of yen in terms of transaction profit and loss. This transaction result may be the result of the actual transaction, but is not limited to this, and may be composed of the result of the virtual trade performed virtually.

In the example of FIG. 3, it is assumed that the input data is, for example, reference algorithm information P01 to P03. The reference algorithm information P01 to P03 as such input data is linked to the output. In this output, the transaction result as the output solution is displayed.

The reference algorithm information is associated with each other through three or more levels of association with the transaction performance as the output solution. The reference algorithm information is arranged on the left side via this degree of association, and each transaction result is arranged on the right side via this degree of association. The degree of association indicates which transaction performance is highly relevant to the reference algorithm information arranged on the left side. In other words, this degree of association is an index indicating what kind of transaction performance each reference algorithm information is likely to be associated with, and is used to select the most probable transaction result from the reference algorithm information. It shows the accuracy. In the example of FIG. 3, w13 to w19 are shown as the degree of association. These w13 to w19 are shown in 10 stages as shown in Table 1 below, and the closer to 10 points, the higher the degree of relevance of each combination as an intermediate node to the transaction performance as an output. On the contrary, the closer to one point, the lower the degree of relevance of each combination as an intermediate node to the transaction performance as an output.

The search device 2 acquires in advance the degree of association w13 to w19 of three or more stages shown in FIG. That is, the search device 2 accumulates past data as to which of the reference algorithm information and the transaction result in that case is more in determining the actual search solution, and analyzes and analyzes these. The degree of association shown in FIG. 3 is created in.

For example, it is assumed that +400,000 yen is overwhelmingly large as a result of conducting a transaction for a certain reference algorithm information P01 for a predetermined period. As a result of conducting a transaction for the same reference algorithm information P01 for a predetermined period, it is assumed that -500,000 yen is small. In such a case, the degree of association of 400,000 yen becomes stronger and the degree of association of −500,000 yen becomes lower.
Incidentally, this predetermined period is not limited to the same period, and the transaction results may be calculated for different periods.
That is, in the formation of this degree of association, automatic trading is performed based on the reference algorithm information from the chart of the past stock price, and the transaction result including the transaction profit / loss or the winning rate based on the automatic trading in the predetermined period is acquired. By using these reference algorithm information and the data set of transaction results, the degree of association is formed as described above.

This analysis and analysis may be performed by artificial intelligence. In such a case, for example, in the case of reference algorithm information P01, analysis is performed from the data of each past transaction result. In the case of the reference algorithm information P01, if there are many cases of transaction result A1 (+1.2 million yen), the degree of association leading to this transaction result A1 is set higher, and the case of transaction result A3 (+400,000 yen) is set. If there are many, the degree of association that leads to this transaction result A3 is set higher. For example, in the example of the reference algorithm information P01, the transaction result A1 and the transaction result A3 are linked. The degree of association is set to 2 points.

Further, the degree of association shown in FIG. 3 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

Such a degree of association is what is called learned data in artificial intelligence. After creating such trained data, the above-mentioned trained data will be used in actually extracting the optimum algorithm from now on. In such a case, first, among the transaction results, the better transaction results are preferentially extracted. In the case of the example of FIG. 3, the transaction results are in the order of A1, A4, A3, A2. In such a case, only the best transaction result A1 may be extracted, or if the above two cases, A1 and A4 may be extracted. If the better transaction results are preferentially extracted, some may be extracted in the order of superiority without being limited to the best ones.

For the transaction results extracted in this way, a reference automatic trading algorithm with a higher degree of relevance is searched for. For example, in the case of the example of FIG. 3, it is assumed that the reference algorithm information P01 is associated with w13 and the reference algorithm information P02 is associated with w14'with respect to A1 having the best transaction result. At this time, when the degree of association w14'is larger than the degree of association w13 with respect to A1, it is suggested that the reference algorithm information P02 is most likely to realize this transaction result A1. .. In such a case, the reference algorithm information P02 is displayed as a proposal candidate.

Of course, the present invention is not limited to displaying only the reference algorithm information that leads to the highest degree of association, and the reference algorithm information that leads to the next highest degree of association may be displayed, and then the reference algorithm information that leads to the next highest degree of association may be displayed. Reference algorithm information that leads to a high degree of association may be displayed. That is, the proposed display of the reference algorithm may be displayed with priority given to the reference algorithm information having a higher degree of association with the transaction results.

In this way, it is possible to search the reference algorithm information that can be linked to the highest transaction performance from the newly acquired market condition information through the past transaction results and display it to the user (consultant). By looking at the results of this proposal, the user (consultant) can know that there is a high possibility that the profit in the trade can be further improved by selecting the searched reference algorithm.

Further, the degree of association shown in FIG. 3 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

In such a case, as shown in FIG. 4, reference algorithm information is input as input data, data of each transaction result is output as output data, and at least one or more hidden layers are formed between the input node and the output node. It may be provided and machine-learned. On the contrary, the transaction result may be input and the reference algorithm information may be output. In particular, in the present invention, since the one with high transaction results is preferentially input and the reference algorithm information that can be realized with a higher probability is output, the node on the right side is the input and the node on the left side is the input. In some cases, the node of is output.

The above-mentioned degree of association is set in either one or both of the input node and the hidden layer node, and this is the weight of each node, and the output is selected based on this. Then, when the degree of association exceeds a certain threshold value, the output may be selected.

In the present invention, the market condition information may be newly taken in, and the reference algorithm information which may have better transaction results may be proposed based on the taken-in market condition information.

In such a case, first, a chart of stock prices at the time of new stock trading is acquired, and market information composed by applying this to a chart pattern categorized in advance is acquired.

FIGS. 5 and 6 show an example of a chart pattern of trading signals of the chart. For example, in FIG. 5A, when the stock price repeatedly moves up and down with respect to the moving average line, the buying signal is when the stock price falls to the moving average line. Further, in FIG. 5B, a buy signal is obtained when the stock price breaks out of the upside resistance line after the fraying market continues for a long time. FIG. 5C is said to be a W bottom type, and when the stock price hits the low price twice in the low price range, it becomes a buy signal. FIG. 5 (d) is called the head and shoulders shoulders, which is a chart pattern in which the stock price hits the low price three times in the low price range, and the middle is the cheapest, and when this appears, it becomes a buy signal. FIG. 6A shows a case where the stock price soars and then immediately plummets, and a long lower beard candlestick or the Taiyo line appears and reverses. When this appears, it is a sign of buying. Fig. 6 (b) is said to be the bright star at the end of Mikawa. When the shape of the Taiyo line appears, it will be a signal to buy in the near future. FIG. 6 (c) is said to be Mikawa Akegarasu, and represents a V-shaped conversion with the red three soldiers after plunging with the black three soldiers (three feathers), and is a sign of buying. In Fig. 6 (d), the bright star of Mikawa evening becomes the Taiyo line in the ascending phase, and the upper mad is opened and the beard and the substance are short. It will be a sign of a downturn and a sell signal.

Such signals are born from past empirical rules in stock trading, but in the present invention, the market information and the reference market information may be applied to the chart pattern type of these trading signals. ..

For this fitting, a determination model generated by machine learning as shown in FIG. 7 may be used. In this determination model, an image of a chart pattern of a trading signal consisting of the above-mentioned example is used as teacher data. The input is a chart of each stock price, and the output is a chart pattern type. When the chart is acquired, fitting is performed based on the judgment model generated by this machine learning, and it is determined what type of trading signal is applied to the chart pattern.

For example, as a result of inputting a chart, it is possible to determine whether or not the output corresponds to the type of trading signal such as Akegarasu Mikawa and Akegarasu Mikawa evening.

As a specific process in FIG. 7, a chart of each stock price showing the market conditions is acquired. Then, it is determined what kind of chart pattern the acquired chart applies to through the determination model shown in FIG. 7, and each reference market condition information is applied to the type of chart pattern. As a result, this reference market information will be represented by a typified chart pattern. Market information will also be represented by a similarly typified chart pattern.

Next, even when market condition information is acquired, it is determined through the determination model shown in FIG. 7 what type of chart pattern the acquired chart applies to, and each reference market condition information is applied to the type of chart pattern. go. As a result, this market information will be represented by a typified chart pattern. The type of chart pattern of such market information is determined through the above-mentioned degree of association to determine what type of chart pattern of the reference market information is applicable.

The chart pattern is not limited to the one corresponding to the trading signal such as Akegarasu Mikawa and Akegarasu Mikawa evening described above, and any other chart may be patterned. Some charts move irregularly, and some charts have almost no change in candlesticks, but such charts may be incorporated into the chart pattern as one pattern. This chart pattern may be configured in seconds, minutes, or hours, or may be configured in units of one day, one week, or one month.

As shown in FIG. 8, when the chart of the past stock price is acquired and the reference market condition information R1 which is applied to the chart pattern Q categorized in advance is acquired, it is applied to the reference market condition information R1. Automatic trading is performed on the chart pattern Q based on the reference algorithm information P01 to P03. This automatic trade may be virtual. As a result, the transaction results are output for each of the reference algorithm information P01 to P03. The above-mentioned degree of association is similarly formed through the data set of the reference algorithm information P01 to P03 and the transaction result.

As a result, the degree of association between the reference algorithm information and the transaction result is formed for each reference market condition information R1, R2, ... Rn.

Next, a chart of stock prices at the time of new stock trading is acquired, and market information composed by applying this to a chart pattern categorized in advance is acquired. The chart of the stock price to be acquired is the chart most recently at the timing of new stock trading, and the period is the same as that at the time of acquisition of the reference market information.

It is determined what kind of reference market information the market information obtained in this way corresponds to. In this determination, the correspondence may be determined based on the similarity between the chart pattern constituting the market condition information and the chart pattern constituting the reference market condition information.

For example, in the example of FIG. 8, it is assumed that the newly acquired chart pattern Q'of the market condition information R1'is similar to the chart pattern Q of the reference market information R1. In such a case, it is determined that the market condition information R1'and the reference market condition information R1 are in a corresponding relationship with each other, and the chart pattern Q of the reference market condition information R1 is actually subjected to the reference algorithms P01 to P03. Refer to the degree of association for which automatic transactions were conducted. Then, among the transaction results related to the degree of association, the better transaction results are preferentially extracted, and the subsequent process is the same as described above.

This makes it possible to propose an optimal reference algorithm based on the latest chart of the timing at which a transaction is to be made.

The similarity between the chart pattern of the market information and the chart pattern of the reference market information may be discriminated based on the feature amount of the imaged chart by using, for example, a deep learning technique.

Further, the present invention is not limited to the above-described embodiment. As shown in FIG. 9, the above-mentioned degree of association is such that any one or more of the reference event information, the reference external environment information, the reference expert opinion information, and the reference natural environment information are related. You may.

The reference event information here refers to various social news, events, incidents, celebrations, celebrations, etc. that occurred in Japan or abroad, as well as news, events, incidents, celebrations, celebrations, etc. that occurred about each company. It is a concept that includes. This reference event information can be obtained from blogs, analyst reports, securities reports, advertisements, press releases, news articles, etc. regarding each company or society as a whole.

External environmental information for reference is politics and economy outside the company, such as GDP, employment statistics, mining and manufacturing production index, capital investment, labor force survey, business conditions index, consumption expenditure, new car sales volume, consumer price index, etc. Includes various data on, society, technology, etc.

Reference expert opinion information means any information that gives an expert opinion on the increase or decrease in stocks, exchange rates, futures, etc., and is the stock price forecast and stocks published in analyst reports and newspaper articles. Expert comments and views on the reasons for the increase or decrease. In addition, the reference expert opinion information may simply be the prediction itself as to whether the stock price of each stock will rise, fall, or remain unchanged. This reference expert opinion information includes opinions on the Nikkei Stock Average as a whole, opinions on specific segments and industries, and opinions on individual stocks. In addition, the reference expert opinion information may include comments by experts (analysts) posted on the Internet and forecasts of increase or decrease.

Reference natural environment information means all information related to the natural environment such as disaster data, temperature data, precipitation data, wind direction data, humidity data, etc., and data on the past natural environment released by the Meteorological Agency, or private companies and private companies. Data on the past natural environment released by individuals.

Hereinafter, the above-mentioned reference event information, reference external environment information, reference expert opinion information, and reference natural environment information may be collectively referred to as reference information.

All of this reference information is configured as a factor that influences the degree of association. For example, if a government official makes a statement that has an impact on the market, it may have a large impact on the stock price chart, and as a result, when automatic trading is performed based on the conventional reference algorithm. , There are cases where it does not necessarily lead to the expected transaction results. In such a case, the degree of association w may be adjusted. Especially in the case of reference information that is negative to the market, the degree of association w that leads to better transaction results is lowered, and in the case of reference information that is positive to the market, better transaction results are obtained. It may be set to increase the degree of connection w.

The content of such reference information and the rule of interlocking with the degree of association with the content are determined in advance. Then, at the stage of actually performing the solution search, the information corresponding to the reference information is acquired. As an example of the information corresponding to this reference information, the event information is acquired for the reference event information, the external environment information is acquired for the reference external environment information, and the reference expert opinion information is used. For that, the expert opinion information is acquired, and for the reference natural environment information, the natural environment information is acquired. The type of event information is the same as the reference event information, the type of external environment information is the same as the reference external environment information, and the type of expert opinion information is the same as the reference expert opinion information. , The types of natural environment information are the same as those for reference natural environment information.

Then, after referring to the reference information corresponding to the newly acquired information, the degree of association is actually adjusted. For example, if the newly acquired event information is negative in the market, the corresponding reference event information (negative for the market) should originally lower the degree of association w that leads to better transaction results. It has been set. Therefore, the degree of association w is lowered via the reference event information corresponding to this event information. As a result, the required search solution changes according to the newly acquired event information.

That is, after referring to the reference information (reference event information, etc.) according to the acquired information (event information, etc.), the automatic trading algorithm having a higher degree of association is prioritized and displayed.

At this time, as shown in FIG. 10, the degree of association may be configured by combining the reference algorithm information and the reference information (reference event information). The degree of association of reference information other than the reference event information may be combined in the same manner, but the following describes the reference event information as an example.

In the example of FIG. 10, it is assumed that the input data is, for example, reference algorithm information P01 to P03 and reference event information P14 to 17. The intermediate node shown in FIG. 4 is a combination of reference event information and reference algorithm information as such input data. Each intermediate node is further linked to the output. This output is the transaction performance.

Each combination (intermediate node) of the reference algorithm information and the reference event information is related to each other through three or more levels of association with the transaction performance as this output solution. Reference algorithm information and reference event information are arranged on the left side via this degree of association, and each transaction result is arranged on the right side via this degree of association. The degree of association indicates the degree of relevance to each transaction result with respect to the reference algorithm information and the reference event information arranged on the left side. In other words, this degree of association is an index indicating what kind of transaction results each reference algorithm information and reference event information are likely to be associated with, and is the most from the reference algorithm information and the reference event information. It shows the accuracy in selecting a reliable transaction result.

In the example of FIG. 10, w13 to w22 are shown as the degree of association. As shown in Table 1, these w13 to w22 are shown in 10 stages, and the closer to 10 points, the higher the degree of relevance of each combination as an intermediate node to the output, and conversely, 1 point. The closer they are, the less relevant each combination as an intermediate node is to the output.

The search device 2 acquires in advance the degree of association w13 to w22 of three or more stages shown in FIG. That is, the search device 2 obtains past data as to which of the reference algorithm information, the reference event information, and the stock price increase / decrease data of each stock in that case is suitable for determining the actual search solution. By accumulating and analyzing and analyzing these, the degree of association shown in FIG. 10 is created.

This analysis and analysis may be performed by artificial intelligence. In such a case, for example, when the reference algorithm information P01 is the reference event information P16, the increase / decrease data of the stock price of each stock is analyzed from the past data. If each transaction result has many cases of transaction result A1, the degree of association leading to this transaction result A1 is set higher, and if there are many cases of transaction result A2 and few cases of transaction result A1, the transaction result is set higher. The degree of association that leads to A2 is set high, and the degree of association that leads to transaction performance A1 is set low. For example, in the example of the intermediate node 61a, it is linked to the output of the transaction result A1 and the transaction result A2. The degree is set to 2 points.

Further, the degree of association shown in FIG. 10 may be composed of the nodes of the neural network in artificial intelligence. That is, the weighting coefficient for the output of the node of this neural network corresponds to the above-mentioned degree of association. Further, the network is not limited to a neural network, and may be composed of all decision-making factors constituting artificial intelligence.

In the example of the degree of association shown in FIG. 10, the node 61b is a node that combines the reference algorithm information P01 with the reference event information P14, and the degree of association of the transaction result A3 is w15 and the degree of association of the transaction result A5. Is w16. The node 61c is a node that is a combination of the reference event information P15 and P17 with respect to the reference algorithm information P02, and the degree of association of the transaction result A2 is w17 and the degree of association of the transaction result A4 is w18.

Such a degree of association is what is called learned data in artificial intelligence. After creating such learned data, when actually performing a search for proposing an automatic trading algorithm from now on, the above-mentioned learned data will be used.

In addition, Table 2 below shows examples of the degrees of association w1 to w12 extending from the input.

The intermediate node 61 may be selected based on the degree of association w1 to w12 extending from this input. That is, the larger the degree of association w1 to w12, the heavier the weighting in the selection of the intermediate node 61 may be. However, the degrees of association w1 to w12 may all have the same value, and the weights in the selection of the intermediate node 61 may all be the same.

By the way, this reference event information constitutes the intermediate node 61 in combination with the reference algorithm information, but is not limited to this, and is not limited to this, and is the intermediate described above in combination with the transaction result and the reference event information. Node 61 may be configured.

That is, the transaction result and the reference event information are input, and the output is the reference algorithm information, which constitutes the degree of association. Then, the input that is particularly excellent in the transaction results is the event information, and the output is the reference algorithm information.

It is possible to search for a solution by configuring the degree of association for reference information other than the reference event information in the same manner.

In the above-described embodiment, the automatic trading of stock transactions has been described as an example, but the present invention is not limited to this, and the present invention may be applied to the automatic trading of foreign exchange (FX) and futures. Of course.

In the above-mentioned degree of association, the degree of association is expressed by a 10-step evaluation, but it is not limited to this, and it may be expressed by a degree of association of 3 or more levels, and conversely, it may be expressed by 3 or more levels. For example, 100 steps or 1000 steps may be used. On the other hand, this degree of association does not include those expressed in two stages, that is, whether or not they are related to each other, either 1 or 0.

According to the present invention having the above-described configuration, anyone can easily search for algorithmic information suitable for performing automatic transactions without any special skill or experience. Further, according to the present invention, it is possible to determine the search solution with higher accuracy than that performed by a human being. Further, by configuring the above-mentioned degree of association with artificial intelligence (neural network or the like), it is possible to further improve the discrimination accuracy by learning this.

Since there are many cases where the input data and the output data described above do not exist exactly the same in the process of learning, the input data and the output data may be classified by type. That is, the information P01, P02, ... P15, 16, ... That constitute the input data are classified according to the criteria classified in advance on the system side or the user side according to the content of the information, and the classified inputs. A data set may be created between the data and the output data and trained.

Further, according to the present invention, there is a feature that an optimum solution search is performed through the degree of association set in three or more stages. The degree of association can be described by, for example, a numerical value from 0 to 100% in addition to the above-mentioned 10 steps, but the degree of association is not limited to this, and any step can be described as long as it can be described by a numerical value of 3 or more steps. It may be configured.

In a situation where there are multiple possible candidates for a search solution by discriminating stocks with higher profit margins and lower risk based on the degree of association expressed by such numerical values of three or more levels. It is also possible to search and display in descending order of the degree of association.

In addition to this, according to the present invention, it is possible to judge without overlooking the discrimination result of the output having an extremely low degree of association such as 1%. It warns the user that even a judgment result with an extremely low degree of association is connected as a slight sign and may be useful as the judgment result once every tens or hundreds of times. be able to.

Further, according to the present invention, there is an advantage that the search policy can be determined by the method of setting the threshold value by performing the search based on the degree of association of three or more stages. If the threshold value is lowered, even if the above-mentioned degree of association is 1%, it can be picked up without omission, but it is unlikely that a more appropriate discrimination result can be detected favorably, and a lot of noise may be picked up. be. On the other hand, if the threshold value is raised, there is a high possibility that the optimum search solution can be detected with a high probability. Sometimes the solution is overlooked. It is possible to decide which one to prioritize based on the ideas of the user side and the system side, but it is possible to increase the degree of freedom in selecting the points to be emphasized.

Further, in the present invention, the above-mentioned degree of association may be updated. This update may reflect information provided via a public communication network such as the Internet. In addition to the reference algorithm information, when knowledge, information, and data related to various event information, external environment information, etc. are acquired, the degree of association is increased or decreased according to these.

In other words, this update corresponds to learning in the sense of artificial intelligence. It can be said that it is a learning act because it acquires new data and reflects it in the learned data.

In addition, this update of the degree of association is not based on the information that can be obtained from the public communication network, but is also updated by the system side or the user side based on the contents of research data, treatises, conference presentations, newspaper articles, books, etc. by experts. It may be updated artificially or automatically. Artificial intelligence may be utilized in these update processes.

Further, the process of first creating the trained model and the above-mentioned update may use not only supervised learning but also unsupervised learning, deep learning, reinforcement learning, and the like. In the case of unsupervised learning, instead of reading and learning the data set of input data and output data, information corresponding to the input data is read and trained, and the degree of association related to the output data is self-formed from there. You may let it.

1 Automatic trade algorithm proposal system 2 Search device 21 Internal bus 23 Display unit 24 Control unit 25 Operation unit 26 Communication unit 27 Estimate unit 28 Storage unit 61 Node

Claims (10)

In the automatic trading algorithm proposal program that proposes an automatic trading algorithm that automatically trades stocks
Three or more levels of association between each reference algorithm information for each automatic trading algorithm that automatically trades stocks and the transaction results that were automatically traded based on the above reference algorithm information from the acquired stock price chart. The association degree acquisition step to be acquired in advance and
Among the above-mentioned transaction results related to the degree of association acquired in the above-mentioned degree of association acquisition step, the transaction result extraction step for preferentially extracting the better transaction results, and the transaction result extraction step.
An automatic trading algorithm proposal program characterized by causing a computer to execute a display step in which a reference automatic trading algorithm having a higher degree of association with the transaction results extracted by the above transaction result extraction step is prioritized and displayed. In the above-mentioned linkage degree acquisition step, automatic trading is performed based on the above-mentioned reference algorithm information from a chart of past stock prices, and transaction results including transaction profit / loss or winning rate based on the automatic trading in a predetermined period are acquired and used for the reference. The automatic trading algorithm proposal program according to claim 1, wherein the above-mentioned degree of association is formed based on the algorithm information and the acquired transaction results. It has an information acquisition step to acquire market information composed by acquiring a stock price chart at the time of new stock trading and applying it to a chart pattern categorized in advance.
In the above-mentioned degree of association acquisition step, a chart of past stock prices is acquired, and the reference market condition information applied to the chart pattern categorized in advance and the chart pattern applied to the reference market condition information are used for the above reference. In addition to performing automatic trading based on algorithm information, transaction results including transaction profit / loss or win rate based on the automatic trade in a predetermined period are acquired, and the above-mentioned degree of association is formed based on the reference algorithm information and the acquired transaction results. ,
In the transaction result extraction step, the above-mentioned transaction related to the above-mentioned degree of association is automatically traded for the chart pattern applied to the reference market condition information corresponding to the above-mentioned market condition information consisting of the types of chart patterns applied in the above-mentioned information acquisition step. The automatic trading algorithm proposal program according to claim 1 or 2, wherein the better transaction results are preferentially extracted from the results. In the information acquisition step and the association degree acquisition step, the image of the chart is used as the teacher data, the input is each of the acquired charts, the output is a categorized chart pattern, and the judgment model generated by machine learning is used. The automatic trading algorithm proposal program according to claim 3, wherein the above fitting is performed. In the above-mentioned linkage degree acquisition step, the above-mentioned association degree related to the reference event information reflecting the event that occurred during the period of the acquired stock price chart is acquired.
In the above information acquisition step, event information that reflects the event that occurred at the time of new stock trading is acquired.
In the display step, after referring to the reference event information corresponding to the event information acquired in the information acquisition step, priority is given to an automatic trading algorithm having a higher degree of association with the transaction results extracted by the transaction result extraction step. The automatic trading algorithm proposal program according to any one of claims 1 to 4, wherein the automatic trading algorithm is displayed. In the above-mentioned linkage degree acquisition step, the above-mentioned association degree related to the reference external environment information reflecting the external environment in the period of the acquired stock price chart is acquired.
In the above information acquisition step, external environment information that reflects the external environment at the time of new stock trading is acquired.
In the display step, after referring to the reference external environment information corresponding to the external environment information acquired in the information acquisition step, an automatic trading algorithm having a higher degree of association with the transaction results extracted by the transaction result extraction step is used. The automatic trading algorithm proposal program according to any one of claims 1 to 4, which is characterized in that it is displayed with priority. In the above-mentioned degree of association acquisition step, the above-mentioned degree of association is acquired with reference expert opinion information reflecting the expert's opinion announced during the period of the acquired stock price chart.
In the above information acquisition step, expert opinion information that reflects the expert opinions announced at the time of new stock trading is acquired.
In the display step, after referring to the expert opinion information for reference according to the expert opinion information acquired in the information acquisition step, the automatic trade having a higher degree of association with the transaction results extracted by the transaction result extraction step. The automatic trading algorithm proposal program according to any one of claims 1 to 4, wherein the algorithm is displayed with priority. In the above-mentioned degree of association acquisition step, the above-mentioned degree of association related to the reference natural environment information reflecting the information on the natural environment during the period of the acquired stock price chart is acquired.
In the above information acquisition step, the natural environment information that reflects the information on the natural environment at the time of new stock trading is acquired.
In the display step, after referring to the reference natural environment information corresponding to the natural environment information acquired in the information acquisition step, an automatic trading algorithm having a higher degree of association with the transaction results extracted by the transaction result extraction step is used. The automatic trading algorithm proposal program according to any one of claims 1 to 4, which is characterized in that it is displayed with priority. In the automatic trading algorithm proposal program that proposes an automatic trading algorithm that automatically performs foreign exchange trading
Three or more levels of association between the reference algorithm information for each automatic trading algorithm that automatically performs forex trading and the transaction results for automatic trading based on each of the above reference algorithms from the acquired foreign exchange chart are set in advance. The degree of association acquisition step to acquire and
Among the above-mentioned transaction results related to the degree of association acquired in the above-mentioned degree of association acquisition step, the transaction result extraction step for preferentially extracting the better transaction results, and the transaction result extraction step.
An automatic trading algorithm proposal program characterized by causing a computer to execute a display step in which a reference automatic trading algorithm having a higher degree of association with the transaction results extracted by the above transaction result extraction step is displayed with priority. In the automatic trading algorithm proposal program that proposes an automatic trading algorithm that automatically trades futures
Three or more levels of association between the reference algorithm information for each automatic trading algorithm that automatically trades futures and the transaction results that were automatically traded based on each of the above reference algorithms from the acquired futures chart are set in advance. The degree of association acquisition step to acquire and
Among the above-mentioned transaction results related to the degree of association acquired in the above-mentioned degree of association acquisition step, the transaction result extraction step for preferentially extracting the better transaction results, and the transaction result extraction step.
An automatic trading algorithm proposal program characterized by causing a computer to execute a display step in which a reference automatic trading algorithm having a higher degree of association with the transaction results extracted by the above transaction result extraction step is prioritized and displayed.

Images