TWM645669U - Stock price factor analysis system - Google Patents

Abstract

A stock price factor analysis system includes a model training module, a SHAP value calculation module, and a determination module. The model training module is configured to train a model by a machine learning method based on multiple factor data corresponding to each of multiple factors affecting stock prices and historical return rates or maximum fallback values of a preset time period in historical stock prices. The SHAP value calculation module is electrically connected to the model training module and configured to substitute the factor data into the model to obtain respective SHAP values corresponding to the factor data. The determination module is electrically connected to the SHAP value calculation module and configured to determine a degree of influence of each of the factors according to the SHAP values corresponding to the factor data, so as to select a plurality of key factors, and then assist investors in determining trading of stock markets.

Description

Stock price factor analysis system

This disclosure document relates to a stock price factor analysis system, particularly a system for determining key factors affecting stock prices.

There are many factors that affect the stock market, such as the rise or fall of interest rates, the cyclical cycle of prosperity due to supply and demand, company revenue, even natural and man-made disasters, government policies, etc., which will cause the stock price to fluctuate up and down. How to predict the stock price trend is The issue that most investors are most interested in. In order to assist investors in judging stock price trends, various methods of predicting stock price changes have also emerged, such as making judgments based on historical prices, price and volume strength, technical aspects, fundamentals and/or chips, etc.

However, the stock market is changing rapidly, and it is very difficult to accurately predict the future rise and fall of stock prices. However, if you can grasp the key factors of stock price trends, you can still determine the general direction of the stock market's rise and fall and its current position in the boom or product cycle. Therefore, in order to find out the key factors that mainly affect the stock price trend among the many factors that affect the stock market, it is necessary to develop analysis technology on stock price factors to further assist investors in judging stock market buying and selling.

A stock price factor is proposed in one of the technical aspects of this disclosure document. element analysis system. The stock price factor analysis system includes a model training module, a SHAP value calculation module and a judgment module. The model training module is used to train the model using a machine learning method based on the plurality of factor data corresponding to each of the plurality of factors that affect the stock price and the historical return rate or maximum retracement value of a preset time period in the historical stock price. The SHAP value calculation module is electrically connected to the model training module, and is used to substitute the factor data into the model, and perform SHAP algorithm processing on the model to obtain the SHAP values corresponding to the factor data. The judgment module is electrically connected to the SHAP value calculation module, and is used to judge the degree of influence of each of these factors based on the SHAP values corresponding to the data of these factors, so as to select a plurality of key factors.

In one embodiment, the SHAP value corresponding to each of these factor data is the sum or average of a plurality of SHAP values corresponding to a plurality of time intervals in the preset time period for each of these factor data.

In one embodiment, the stock price factor analysis system further includes a historical data calculation module. The historical data calculation module is used to calculate the historical return rate of this preset time period based on historical stock prices for access by the model training module.

In one embodiment, the stock price factor analysis system further includes a numerical conversion module. The numerical conversion module is used to convert the historical return rate or the maximum fallback value into a numerical value. The model training module is based on these factor data and the numerically converted historical return rate or maximum fallback value, and trains the historical return rate or maximum fallback value using machine learning methods. Model.

In one embodiment, the stock price factor analysis system further includes a data collection module. The data collection module is used to select from the group consisting of overall economic indicators, fundamental indicators, raw material indicators, chip indicators, foreign exchange, and technical indicators. Select these factors and collect factor data corresponding to each of these factors for access by the model training module.

In one embodiment, the factor data is the corresponding growth rate data of these factors or data generated by numerical conversion of these factors.

In one embodiment, the model training module trains the model using automated machine learning methods.

In one embodiment, the model training module divides these factor data into a training set, a verification set, and a test set in time sequence, and based on the training set, verification set, test set, and historical return rate or maximum fallback value, the machine Learning methods train, validate, and test this model.

In one embodiment, the stock price factor analysis system further includes a model database. The model database is used to store models trained by the model training module.

In one embodiment, the stock price factor analysis system further includes a status monitoring module. The status monitoring module is used to display model information in the model database on the monitor.

100, 200: Stock price factor analysis method

S110, S120, S130, S140, S210, S220, S232, S234, S240: Steps

400, 500: Stock price factor analysis system

410, 510: Data collection module

420, 520: Influencing factors database

430, 530: Historical data calculation module

440, 540: Historical data database

450, 550: Model training module

460, 560: SHAP value calculation module

470, 570: Judgment module

480, 580: Model database

490, 590: Status monitoring module

512: The first numerical conversion module

532: Second numerical value conversion module

600:Server

610a, 610b: User equipment

Figure 1 is a flow chart of a stock price factor analysis method according to one embodiment of this disclosure document.

Figure 2 is a flow chart of a stock price factor analysis method according to one embodiment of this disclosure document.

FIG. 3A is an example diagram of key factor ranking results according to an embodiment of the present disclosure.

FIG. 3B is an example diagram of key factor ranking results according to an embodiment of the present disclosure.

FIG. 3C is an example diagram of key factor ranking results according to an embodiment of the present disclosure.

Figure 4 is a schematic diagram of the stock price factor analysis system architecture according to one embodiment of this disclosure document.

Figure 5 is a schematic diagram of the stock price factor analysis system architecture according to one embodiment of this disclosure document.

Figure 6 is a usage scenario diagram of the stock price factor analysis system according to one embodiment of this disclosure document.

The following is a detailed description of the embodiments in conjunction with the accompanying drawings. However, the specific embodiments described are only used to explain the present invention and are not used to limit the present invention. The description of the structural operations is not intended to limit the order of its execution. Any The structure of the components being reassembled to produce a device with equal functions is within the scope of the disclosure of the present invention.

In addition, the terms "first" and "second" mentioned in the present invention do not represent any order, quantity or importance, but are only used to distinguish different parts, and the drawings are only used for schematic illustration.

Unless otherwise noted, the terms used throughout the specification and patent claims generally have their ordinary meanings as used in the field, in the disclosure and in the particular content. Certain terms used to describe the present disclosure are discussed below or elsewhere in this specification to provide an understanding of Those skilled in the art are provided with additional guidance regarding the description of this disclosure.

Please refer to FIG. 1 , which illustrates a flow chart of a stock price factor analysis method 100 according to an embodiment of this disclosure document. The stock price factor analysis method 100 is used to determine one or more key factors from many factors that affect the stock price. It includes steps S110, S120, S130 and S140, and the contents of each step are as follows:

S110: Collect multiple factor data corresponding to multiple factors that affect the stock price, and calculate the historical return rate or maximum drawdown value (Max drawdown, MDD) of a preset time period in the historical stock price.

S120: Use machine learning methods to train the model based on factor data and historical return rates or maximum drawdown values.

S130: Execute SHAP algorithm (SHapley Additive exPlanations) processing to obtain SHAP values corresponding to each factor data.

S140: Determine the degree of influence of each factor data based on its corresponding SHAP value to select multiple key factors.

Specifically, in step S110, multiple factors corresponding to multiple factors that affect the stock price are collected. For example, for a certain stock product A, factors that may affect its stock price at a certain point in time or at a certain time are collected. The data corresponding to the segment. Among them, the processor can automatically collect the corresponding factor data of the factors that affect the stock price and store it in the database. Alternatively, the user can also input the factors and the corresponding factor data themselves and store them in the database. The factors that affect the stock price are selected from the overall economic indicators, fundamental indicators, raw material indicators, chip indicators, foreign exchange, and technical indicators. group consisting of targets.

Among them, general economic indicators include, for example, the panic index, 10-year Treasury bond rate, 10-year minus 2-year Treasury bond spread, U.S. 30-year mortgage rate, U.S. high yield index option-adjusted spread, U.S. overall inventories, U.S. New orders for manufacturing durable goods, U.S. consumer spending on durable goods, U.S. initial jobless claims, U.S. non-farm payrolls, and U.S. unemployment within five weeks, etc.

Fundamental indicators include, for example, announced or seasonally adjusted monthly revenue, the ratio of monthly revenue to stock price, price-to-earnings ratio calculated from past financial reports, price-to-earnings ratio calculated using announced financial reports and revenue estimates, and stock price calculated from past financial reports. Net worth ratio, stock price to net worth ratio using announced financial reports and revenue estimates, cash yields of announced revenue estimates, and past or upcoming dividend yields, etc.

Raw material indicators include, for example, Brent crude oil, WTI crude oil, natural gas, gold, silver, copper, iron, tin, US soybeans, US wheat, US corn, and CRB commodity index, etc.

Chip indicators include, for example, the average number of chips held by shareholders, the proportion of shareholders holding more than 600 chips, the proportion of shareholders holding more than 1,000 chips, foreign investment trading excess (including the average of 5 days, 20 days, 60 days, etc.), investment trust trading excess (including 5-day, 20-day, 60-day average, etc.), dealer trading excess (including 5-day, 20-day, 60-day average, etc.), financing balance (including 5-day, 20-day, 60-day average), and Securities lending balance (including 5-day, 20-day, 60-day average), etc.

Foreign exchange includes, for example, the U.S. dollar index, the U.S. dollar to Taiwan dollar exchange rate, and the Australian dollar to U.S. dollar exchange rate, etc.

Technical indicators include, for example, trading volume (including 5-day, 20-day, 60-day averages, etc.) and past prices, etc.

It should be understood that the factor data collected by the stock price factor analysis method 100 of the present invention is not limited to the above examples, and any factors that may affect the stock price can be applied to the stock price factor analysis method 100 disclosed by the present invention.

In one embodiment, the factor data corresponding to the factors that affect the stock price may be, for example, data such as the growth rate of each factor at a fixed time interval. For example, according to user needs, the growth rate of various factors related to stock product A in any time period in the past can be collected. In another embodiment, the factor data corresponding to the factors that affect the stock price can also be converted into numerical values such as log values according to actual needs. The present invention does not impose any restrictions on this.

In addition, in step S110, in collecting the historical return rate or the maximum retracement value of the preset time period in the historical stock price, for example, for this stock product A, its return rate or the maximum retracement value of the preset time period in the past is collected. For example, the stock price factor analysis method 100 can collect the historical return rate or maximum drawdown value of stock product A in the past years, months, weeks, or any time interval. Preferably, in step S110 of the stock price factor analysis method 100, the rate of return or the maximum fallback value is collected for a preset period of time after the time point of the aforementioned factor data. For example, when the factor data collected is the data of each influencing factor at time point T1, then the historical return rate or maximum drop value for a preset period after time point T1 is collected.

Next, in step S120, the factor data and historical return rate or maximum retracement value collected in step S110 are used to train the model through machine learning methods. Specifically, the factor data is used as the input of the training model, and the historical return rate or maximum retracement value is used as the target of the training model. Through machine learning Learn to train the most suitable model. Among them, the training model is, for example, XGBoost, Random Forest, neural network, etc., which is not limited by the present invention.

After training the model, in step S130, the stock price factor analysis method 100 substitutes the current factor data or the specific time point to be analyzed into the trained model, and executes the SHAP algorithm (SHapley Additive exPlanations) on the trained model. ) processing to obtain the SHAP value corresponding to each factor data. Among them, the SHAP algorithm can measure the degree of positive or negative contribution of each feature (i.e., factor data) in the model to each prediction (i.e., historical return rate or maximum drawdown value). For example, when the SHAP value is positive, the larger the value, the greater the positive contribution. When the SHAP value is negative, the smaller the value, the greater the negative contribution. Simply put, the greater the absolute value of the SHAP value, the greater the contribution.

Finally, in step S140, the stock price factor analysis method 100 determines the degree of influence/contribution of each factor data on the stock price based on the SHAP value corresponding to the factor data calculated in step S130, so as to select the current or specific time point. Plural key factors.

In another embodiment of the present invention, the criticality of each factor affecting the stock price at a specific period in history can be further analyzed. Please refer to FIG. 2 , which illustrates a flow chart of a stock price factor analysis method 200 according to an embodiment of this disclosure document. Among them, the stock price factor analysis method 200 includes steps S210, S220, S232, S234 and S240. The contents of each step are as follows:

S210: Collect multiple factor data corresponding to multiple factors that affect the stock price, and calculate the historical return rate or maximum drop value of a preset time period in the historical stock price.

S220: Use machine learning methods to train the model based on factor data and historical return rates or maximum drawdown values.

S232: Execute SHAP algorithm processing to calculate a plurality of SHAP values corresponding to a plurality of time intervals in the preset time period for each factor data.

S234: Sum up and/or average a plurality of SHAP values corresponding to each factor data to generate a SHAP Feature Importance (SHAP Feature Importance) corresponding to each factor data.

S240: Based on the SHAP importance values corresponding to the factor data, determine the degree of influence of each factor data to select a plurality of key factors.

Among them, steps S210 and S220 of the stock price factor analysis method 200 correspond to steps S110 and S120 of the stock price factor analysis method 100 in sequence. Please refer to the previous description for relevant descriptions and will not be repeated here.

In this embodiment, the main difference between the stock price factor analysis method 200 and the stock price factor analysis method 100 is that step S130 is further subdivided into steps S232 and S234. Among them, the stock price factor analysis method 200 in step S232 is to substitute the historical return rate or the plurality of time intervals corresponding to the preset time period corresponding to the maximum retracement value stored in step S210 for the model trained in step S220. Factor data, and execute the SHAP algorithm on the model based on each time interval, so the SHAP value corresponding to each factor data in each time interval can be obtained. For example, assuming that the preset time period corresponding to the historical return rate or the maximum retracement value is several years, the preset time period can be divided into a plurality of times based on annual, monthly, weekly, or daily units. interval. It should be understood that the preset time period and time interval can be based on Adjustments are made to actual needs and are not limited by this model.

Next, in step S234, a plurality of SHAP values corresponding to each factor data are added up to obtain SHAP importance values corresponding to each factor data. Preferably, the plurality of SHAP values corresponding to each factor data can be taken as absolute values and then summed to obtain the SHAP importance values corresponding to each factor data. In one embodiment, the SHAP important values can also be further subjected to average calculation or other numerical conversion, which is not limited by the present invention.

Finally, in step S240, the stock price factor analysis method 200 determines the degree of influence of each factor data based on the SHAP importance value corresponding to each factor data, so as to select a plurality of key factors in a specific period.

Please refer to Figures 3A-3C. Figure 3A illustrates an example of the ranking results of the top ten key factors in history according to one embodiment of this disclosure document. Figure 3B illustrates an example of the ranking results of the top five bullish key factors according to one embodiment of this disclosure document. , and Figure 3C illustrates an example diagram of the ranking results of the top five bearish key factors according to one embodiment of this disclosure document.

In the example of FIG. 3A , the stock price factor analysis method 200 is used to determine the criticality ranking of various factors that have historically affected the stock price. The criticality ranking of each factor is based on the size of the SHAP importance value, and the corresponding SHAP importance value is greater. If the factor is larger, the greater the contribution of this factor to the stock product, that is, the more critical the influence ability. For example, in Figure 3A, the SHAP importance value of the factor "U.S. High Yield Index Option Adjusted Spread" is greater than the SHAP importance value of other factors such as "U.S. High Yield Index Option Adjusted Spread YoY", then the stock price factor analysis is used When method 200 is used as the basis for auxiliary judgment, it can be judged The most critical factor is "U.S. High Yield Index Option Adjusted Spread". Therefore, the importance of each factor to stock price changes can be sorted.

It should be understood that Figure 3A is only an example of applying the stock price factor analysis method 200 according to the embodiment of the present invention. It only ranks the top ten key factors in a certain period, and does not list the current ranking of all factors. The selection of factors , the number of key factors and the setting of time can be designed according to the actual needs of the user.

In another embodiment, the analysis can also be performed only on the current rising or falling factors of the stock product. As shown in FIG. 3B , through the stock price factor analysis method 100 , the top five key factors that are bullish (that is, affect the rise of the stock price) can be sorted, for example, according to the current SHAP value. Among them, the key factor for bullishness is to extract the factors with a positive SHAP value (positive contribution) to sort. In the example in Figure 3B, the SHAP values of the top four factors "U.S. overall inventory QoQ", "Financing balance QoQ", "Financing balance (20-day average)" and "U.S. 30-year mortgage interest rate (week) YoY" If both are 0.01, then the criticality of each factor is the same. According to user settings, these factors with equal SHAP values can also be set to the same serial number (for example, both are "1"). This model does not limit this.

In addition, as shown in FIG. 3C , the stock price factor analysis method 100 can also sort the top five key factors that are bearish (that is, affect the stock price to fall) based on the current SHAP value. Among them, the key factor for bearishness is to extract the factors with negative SHAP values (negative contribution) for sorting. As mentioned above, when the SHAP value is negative, the lower the value (or the greater the absolute value), the greater the negative contribution, that is, the more critical it is.

In another embodiment of the present invention, because in the analysis of stock price factors In the process of collecting historical return rates in step S110 of the analysis method 100 or step S210 of the stock price factor analysis method 200, the distribution of historical return rates is not necessarily regular. Therefore, after collecting the historical return rates in steps S110/S210, further The historical return rate is converted into a numerical value, and in steps S120/S220, a machine learning method is used to train the model based on the factor data and the numerically converted historical return rate. The step of converting the historical return rate into numerical value is, for example, converting the historical return rate into numerical conversion such as Yeo-Johnson transformation or Box-Cox transformation. Alternatively, after collecting the maximum fallback value in step S110/S210, the maximum fallback value can be further converted into a numerical value, and in step S120/S220, the model can be trained using a machine learning method based on the factor data and the numerically converted maximum fallback value. . The step of converting the maximum retracement value into a numerical value is, for example, further dividing the maximum retracement value by the highest net value of the stock price product in the preset time period to obtain the maximum retracement value percentage. This can make the training accuracy of the model in steps S120/S220 higher.

In addition, step S120/S220 can also perform model training through automated machine learning (AutoML) to establish an accurate model. In one embodiment, the stock price factor analysis method can also divide the factor data into a training set, a verification set, and a test set in time sequence. Based on the divided training set, verification set, test set and historical return rate or maximum retracement value, the stock price factor analysis method 100/200 can use machine learning methods to train, verify, and test the model in steps S120/S220, thereby avoiding Over-fitting occurs.

Please refer to Figure 4. Figure 4 illustrates an embodiment of this disclosure document. Schematic diagram of the architecture of the stock price factor analysis system 400. The stock price factor analysis system 400 is executed, for example, by a processor, which can execute the stock price factor analysis method 100 and the stock price factor analysis method 200 . The stock price factor analysis system 400 at least includes a data collection module 410, an influencing factor database 420, a historical data calculation module 430, a historical data database 440, a model training module 450, a SHAP value calculation module 460, and a judgment module 470 .

The collection module 410 and the historical data calculation module 430 of the stock price factor analysis system 400 are used to perform steps S110/S210 of the stock price factor analysis method 100/200, for example. The data collection module 410 selects factors that affect the stock price of stock product A from the group consisting of overall economic indicators, fundamental indicators, raw material indicators, chip indicators, foreign exchange, and technical indicators, and collects the respective factors of these factors. The corresponding factor data is provided for storage by the influencing factor database 420. The historical data calculation module 430 calculates the historical return rate or the maximum retracement value of the preset time period based on, for example, the historical stock price of stock product A for storage by the historical data database 440.

The model training module 450 is used to perform, for example, steps S120/220 of the stock price factor analysis method 100/200, which is based on the factor data stored in the influencing factor database 420 and the historical return rate or maximum retracement value stored in the historical data database 440, to Machine learning methods train models. Among them, the model training module 450 can train the model through automated machine learning methods. In addition, the model training module 450 can further divide the factor data into a training set, a verification set, and a test set in time sequence, and based on the training set, the verification set, the test set and the history stored in the historical data database 440 Return rate or maximum drawdown value, train, validate, and test the model using machine learning methods to avoid excessive Over-fitting status.

The SHAP value calculation module 460 is electrically connected to the model training module 450, and is used to perform, for example, step S130 of the stock price factor analysis method 100 and/or steps S232 and S234 of the stock price factor analysis method 200. Among them, when the SHAP value calculation module 460 executes step S130 of the stock price factor analysis method 100, it substitutes the factor data currently or at a specific time point to be analyzed into the model trained by the model training module 450, and performs SHAP calculation on the model. Method processing to obtain the SHAP values corresponding to the factor data, and judge the degree of influence of each of these factors based on the SHAP values corresponding to the factor data, so as to select multiple key factors at present or at a specific time point. Alternatively, when the SHAP value calculation module 460 executes step S232 of the stock price factor analysis method 200, the factor data corresponding to a plurality of time intervals in the preset time period corresponding to the stored historical return rate or the maximum retracement value are substituted respectively into The model training module 450 executes the SHAP algorithm on the trained model based on each time interval to obtain the SHAP value corresponding to each factor data in each time interval. Next, the SHAP value calculation module 460 executes step S234 of the stock price factor analysis method 200 to sum up a plurality of SHAP values corresponding to each factor data to obtain the SHAP importance value corresponding to each factor data.

The judgment module 470 is electrically connected to the SHAP value calculation module 460, and is used to perform steps S140/S240 of the stock price factor analysis method 100/200, for example. It judges each of these factors based on the SHAP value calculated by the SHAP value calculation module 460. to select a plurality of key factors at present or at a specific point in time, or to judge the degree of influence of each of these factors according to the SHAP important value calculated by the SHAP value calculation module 460, so as to Select a plurality of key factors for a specific period. Regarding the data collection module 410, the influencing factor database 420, the historical data calculation module 430, the historical data database 440, the model training module 450, the SHAP value calculation module 460, and the judgment module 470 in the stock price factor analysis system 400 For the detailed step-by-step process of each, please refer to the aforementioned descriptions of the stock price factor analysis method 100 and the stock price factor analysis method 200.

In one embodiment, the stock price factor analysis system 400 further includes a model database 480 . The model database 480 can be used to store one or more models trained by the model training module 450 at each time point for future retrieval when needed. In another embodiment, the stock price factor analysis system 400 may further include a status monitoring module 490 . The status monitoring module 490 can be used to display the information of one or more models stored in the model database 480 on the display for the user to view.

Please refer to FIG. 5 , which is a schematic structural diagram of a stock price factor analysis system 500 according to an embodiment of this disclosure document. The stock price factor analysis system 500 is executed, for example, by a processor, which can also execute the stock price factor analysis method 100 and the stock price factor analysis method 200 . The stock price factor analysis system 500 includes a data collection module 510, a first numerical conversion module 512, an influencing factor database 520, a historical data calculation module 530, a second numerical conversion module 532, a historical data database 540, and a model training module. Group 550, SHAP value calculation module 560, judgment module 570, model database 580, and status monitoring module 590. Among them, the data collection module 510, the influencing factor database 520, the historical data calculation module 530, the historical data database 540, the model training module 550, the SHAP value calculation module 560, and the judgment module in the stock price factor analysis system 500 The group 570, the model database 580, and the status monitoring module 590 correspond in sequence to the data collection module 410, the influencing factor database 420, the historical data calculation module 430, and the historical data database 440 in the stock price factor analysis system 400. , model training module 450, SHAP value calculation module 460, judgment module 470, model database 480, and status monitoring module 490. Their corresponding modules all have the same functions. For relevant instructions, please refer to the previous description. This will not be described again.

Different from the stock price factor analysis system 400, the stock price factor analysis system 500 also has a numerical value conversion module including a first numerical value conversion module 512 and/or a second numerical value conversion module 532. The first numerical conversion module 512 is used to further calculate and convert the factor data collected by the data collection module 510 into growth rate and other data based on a fixed time interval for storage in the influencing factor database 520 . Alternatively, the first numerical conversion module 512 can further convert the factor data into numerical values such as log values for storage in the influencing factor database 520 .

The second numerical conversion module 532 is used to further perform numerical conversion on the historical rate of return calculated by the historical data calculation module 530, such as Yeo-Johnson transformation or Box-Cox transformation, for storage in the historical data database 540. Alternatively, the second numerical conversion module 532 is used to further divide the maximum retracement value calculated by the historical data calculation module 530 by the highest net value of the stock price product in the preset time period to obtain the maximum retracement value percentage for historical data. Database 540 stores. Thereby, the model training module 550 can use the machine learning method to train the model based on the numerically converted factor data and/or the numerically converted historical return rate or maximum drawdown value to obtain a more accurate model.

Please refer to FIG. 6 , which illustrates a usage scenario diagram of the stock price factor analysis systems 400 and 500 according to an embodiment of this disclosure document. In FIG. 6 , the stock price factor analysis systems 400 and 500 can be set up in the server 600 , for example. Server 600 is accessible to one or more user devices. For example, the server 600 can be connected to the user device 610a through wired or wireless access technology for further use of the stock price factor analysis system 400 and/or 500. In addition, the server 600 can also be connected to another user device 610b, for example, through wired or wireless access technology, so that it can further use the stock price factor analysis system 400 and/or 500. Among them, the user equipment can be a computer, a mobile phone, or various non-mobile or portable smart devices, and the user equipment 610a and the user equipment 610b can be the same or different equipment, and can access the server at the same time or not at the same time. 600 for access, which is not limited by this model. It should be understood that the number of user devices in Figure 6 is only for example, and the server 600 can also be connected to more than two user devices to use the stock price factor analysis system 400 and/or 500. The present invention does not not limited.

Although the embodiments of the present invention have been disclosed above, they are not intended to limit the present invention. Anyone skilled in the art can make some modifications and modifications without departing from the spirit and scope of the present invention. Therefore, the protection of the present invention is The scope shall be determined by the appended patent application scope.

400: Stock price factor analysis system

410:Data collection module

420:Influencing factors database

430: Historical data calculation module

440: Historical data database

450:Model training module

460: SHAP value calculation module

470:Judgement module

480:Model database

490:Status monitoring module

Claims (10)

A stock price factor analysis system, set up on a server, including: a model training module, used to base on multiple factors data corresponding to multiple factors that affect stock prices and a historical return in a preset time period in historical stock prices. rate or a maximum fallback value, and trains a model with a machine learning method; a SHAP value calculation module, electrically connected to the model training module, is used to substitute the factor data into the model, and perform SHAP algorithm processing on the model , to obtain the SHAP value corresponding to each of the factor data; and a judgment module, electrically connected to the SHAP value calculation module, for judging the SHAP value of each of the factors according to the SHAP value corresponding to the factor data. degree of influence to select multiple key factors. The stock price factor analysis system as described in request item 1, wherein the SHAP values corresponding to each of the factor data are a plurality of SHAP values corresponding to a plurality of time intervals in the preset time period for each of the factor data. The sum or average of. The stock price factor analysis system described in claim 1 further includes: a historical data calculation module for calculating the historical rate of return or the maximum fallback value of the preset time period based on the historical stock price for training the model. group for access. The stock price factor analysis system as described in request item 1 further includes: A numerical conversion module for converting the historical return rate or the maximum retracement value into a numerical value, wherein the model training module is based on the factor data and the historical return rate or the maximum retracement value after numerical conversion, The model is trained using machine learning methods. The stock price factor analysis system as described in claim 1 further includes: a data collection module for collecting data from a group consisting of overall economic indicators, fundamental indicators, raw material indicators, chip indicators, foreign exchange, and technical indicators. Select the factors and collect the factor data corresponding to each of the factors for access by the model training module. The stock price factor analysis system as described in request item 1, wherein the factor data is the growth rate data corresponding to the factors or data generated by numerical conversion of the factors. The stock price factor analysis system as described in claim 1, wherein the model training module trains the model using an automated machine learning method. The stock price factor analysis system as described in request item 1, wherein the model training module divides the factor data into a training set, a verification set, and a test set in time sequence, and based on the training set, the verification set, Use the test set and the historical return rate or the maximum retracement value to train, verify, and test the model using machine learning methods. The stock price factor analysis system as described in claim 1 further includes: a model database for storing the model trained by the model training module. A stock price factor analysis system as described in claim 9, It also includes: a status monitoring module for displaying the model information in the model database on a monitor.