Security prediction method and device, equipment, medium and program product
By double screening of the institutional returns and fund status of securities issuing institutions and using artificial intelligence models for prediction processing, the limitations of securities screening methods in the existing technology are solved, and the accuracy and stability of securities investment are improved.
Patent Information
- Application Number
- CN202411046409.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-26
AI Technical Summary
The existing securities screening methods have limitations based on fundamentals, capital and technical dimensions, and cannot meet the needs of securities investors, including high time consumption, lagging information updates, ignoring market psychology and short-term fluctuations, and subjectiveness of management quality. Issues such as judgment dependence, difficulty in capturing turning points and emerging opportunities.
By obtaining the securities set to be screened, the screening parameters related to the institutional returns and capital status of the securities issuer are determined, the securities set is double screened, and the screening results are predicted using artificial intelligence models to ensure that the rate of return in the predicted results is greater than that of the forecast results Or equal to the preset threshold.
It realizes high-precision and comprehensive screening of securities collections, reduces the data processing volume of the prediction model, improves the stability and accuracy of the prediction results, and meets the needs of securities investors.
Smart Images

Figure CN119027167A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of securities technology, and in particular to a securities prediction method, device, equipment, medium and program product. Background Art
[0002] In practical applications, securities screening operations are usually performed based on fundamentals, capital and technical dimensions to recommend preferred securities to securities investors. However, although the above methods can reflect valuable market information from different angles, the limitations and defects of the above methods are obvious. Therefore, the preferred securities recommended by the above methods cannot meet the securities investment needs of securities investors. Summary of the invention
[0003] Based on the above technical problems, the embodiments of the present application provide a securities prediction method, device, equipment, medium and program product.
[0004] The technical solution provided by the embodiment of the present application is as follows:
[0005] The present application embodiment first provides a securities prediction method, the method comprising:
[0006] Get the securities collection to be screened;
[0007] Determine a screening parameter; wherein the screening parameter is at least associated with the institutional income of the securities issuing institution and the fund status corresponding to the securities set; the securities issuing institution is associated with the securities set;
[0008] Screening the securities set based on the screening parameters to obtain screening results;
[0009] The screening results are predicted and processed by a prediction model to obtain a prediction result; wherein the yield of the securities included in the prediction result is greater than or equal to a preset threshold.
[0010] In some embodiments, the screening parameters include institutional parameters associated with the institutional earnings of the securities issuer; and determining the screening parameters includes:
[0011] Obtaining an initial securities set associated with the securities issuing institution;
[0012] Determine industry revenue parameters;
[0013] Based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set, the industry return parameter is processed to obtain a processing result;
[0014] Based on the processing results, the mechanism parameters are determined.
[0015] In some embodiments, the processing of the industry return parameter based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set to obtain a processing result includes:
[0016] If the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, the revenue parameter in the industry revenue parameter corresponding to the mth parameter in the first industry parameter is counted to obtain a first result; wherein m and n are both integers greater than or equal to 1;
[0017] Based on the first result, the processing result is determined.
[0018] In some embodiments, the processing of the industry return parameter based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set to obtain a processing result includes:
[0019] If the first industry parameter does not match the second industry parameter, obtain the CSI All-Share Index;
[0020] Determine a second result based on the CSI All-Share Index and the industry return parameter;
[0021] Based on the second result, the processing result is determined.
[0022] In some embodiments, the screening parameters include funding parameters corresponding to the funding status; and determining the screening parameters includes:
[0023] Identify market factors associated with security market conditions;
[0024] The funding parameter is determined based on the market factor; wherein the funding parameter includes a stock factor or a flow factor.
[0025] In some embodiments, the screening parameters include an institutional parameter associated with the institutional income of the securities issuing institution and a capital parameter corresponding to the capital status; the screening of the securities set based on the screening parameters to obtain the screening results includes:
[0026] Performing a first screening on the securities set based on the institutional parameter to obtain a third result;
[0027] A second screening is performed on the third result based on the fund parameter to obtain the screening result.
[0028] In some embodiments, the prediction model includes an artificial intelligence model; and the method further includes:
[0029] Acquire a securities sample; wherein the securities sample includes multiple types of securities and securities returns associated with the multiple types of securities; the number of types of securities included in the securities sample is greater than the number of types of securities included in the securities set;
[0030] Based on the training samples and verification samples in the securities samples, the artificial intelligence model in the initial state is trained to obtain the artificial intelligence model.
[0031] The embodiment of the present application also provides a securities prediction device, the securities prediction device comprising:
[0032] An acquisition module is used to acquire a set of securities to be screened;
[0033] A determination module, configured to determine a screening parameter; screen the securities set based on the screening parameter to obtain a screening result; wherein the screening parameter is at least associated with the institutional income of the securities issuing institution and the capital status corresponding to the securities set; the securities issuing institution is associated with the securities set;
[0034] A prediction module is used to perform prediction processing on the screening results through a prediction model to obtain a prediction result; wherein the yield of the securities included in the prediction result is greater than or equal to a preset threshold.
[0035] An embodiment of the present application also provides an electronic device, which includes a processor and a memory; a computer program is stored in the memory; when the computer program is executed by the processor, it can implement any of the securities prediction methods described above.
[0036] An embodiment of the present application also provides a computer-readable storage medium, in which a computer program is stored; when the computer program is executed by a processor of an electronic device, it can implement any of the above-mentioned securities prediction methods.
[0037] An embodiment of the present application also provides a computer program product, which includes a computer program; when the computer program is executed by a processor of an electronic device, it can implement any of the securities prediction methods described above.
[0038] The securities prediction method provided in the embodiment of the present application obtains a set of securities to be screened, and screens the securities set to obtain a screening result based on screening parameters at least associated with the institutional income of the securities issuing institution and the capital status corresponding to the securities set, and the securities issuing institution is associated with the securities set. In this way, through the above operation, double screening of the securities included in the securities set in the dimensions of securities issuing institution and capital status is achieved, thereby achieving high-precision and comprehensive screening of the securities set; and, the screening results are predicted and processed by the prediction model to obtain the prediction results, thereby reducing the amount of data processing by the prediction model in the above prediction process and improving the efficiency of the prediction model in performing prediction processing on the screening results; on the other hand, since the yield of the securities included in the prediction result is greater than or equal to the preset threshold, combined with the previous steps, the stability and accuracy of the yield of the securities included in the prediction result can be improved, thereby meeting the actual securities investment needs. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] Figure 1 A schematic diagram of a flow chart of a securities prediction method provided in an embodiment of the present application;
[0040] Figure 2 A schematic diagram of a process for screening initial screening results based on funding parameters provided in an embodiment of the present application;
[0041] Figure 3 A schematic diagram of the process of outputting a gold stock combination provided in an embodiment of the present application;
[0042] Figure 4 A schematic diagram of a process for obtaining a gold stock combination provided in an embodiment of the present application;
[0043] Figure 5 A schematic diagram of the model training process provided in the embodiment of the present application;
[0044] Figure 6 A schematic diagram showing the comparative effect of the yield of securities determined by the securities prediction method provided by the embodiment of the present application;
[0045] Figure 7 Another schematic diagram of the comparison effect of the yield of securities determined by the securities prediction method provided by the embodiment of the present application;
[0046] Figure 8 A schematic diagram of the summary effect of the monthly revenue comparison data provided in the embodiment of the present application;
[0047] Fig. 9 A schematic diagram of the structure of a securities prediction device provided in an embodiment of the present application;
[0048] Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0049] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.
[0050] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0051] Accurate security selection is the basic prerequisite for obtaining stable returns in securities trading. Therefore, how to accurately predict or determine securities with rising potential is crucial to improving investors' investment returns.
[0052] Taking stocks as an example, relevant technologies provide technical solutions for selecting stocks from macro fundamentals, capital and technical aspects respectively.
[0053] Among them, the stock selection strategy based on macro fundamental analysis is mainly based on factors such as the company's financial health, industry position, market competitiveness, domestic and foreign macroeconomic data, industrial development status, monetary policy, interest rate level and political situation. The above factors can work together in the securities market, thereby affecting the price of stocks and the yield of bonds. In addition, the core of the macro fundamental stock selection program is to evaluate how the above factors affect the performance of listed companies and the value of stocks. By comprehensively analyzing the above factors, the macro fundamental stock selection program attempts to identify stocks whose intrinsic value is higher than the market price, namely value stocks, and obtain benefits when these stocks are re-evaluated by the market or the price rises. The above stock selection method can meet some of the investment needs of long-term investors. For example, the above stock selection method can meet the needs of investors who seek to obtain sustained returns by investing in stable and profitable companies.
[0054] However, the above stock selection scheme also has the following limitations:
[0055] First, the above stock selection scheme takes a long time and the information update lags significantly. The premise for the implementation of the above stock selection scheme is that it is necessary to deeply analyze and study the company's financial statements, industry status, macroeconomic conditions and other information. The collection and analysis of the above information is usually very time-consuming, and there is a time difference between the release of report data, so that the investment decision determined by the above scheme can only rely on the historical information of the above various information.
[0056] Secondly, the above stock selection scheme ignores the impact of market psychology and short-term fluctuations. The above scheme focuses on evaluating the long-term value of the company and may ignore the impact of market sentiment and short-term price fluctuations on stock performance; in actual application, the irrational behavior of the market in the short term may cause stock prices to deviate significantly from their fundamentals.
[0057] Thirdly, the reliance on subjective judgments of management quality and corporate strategy in the above stock selection schemes may lead to increased uncertainty in investment strategies. The above stock selection methods often rely on the subjective judgments of investors when evaluating the quality of company management and strategic direction, which may lead to increased uncertainty in the analysis results.
[0058] At the same time, the above stock selection schemes are difficult to capture turning points and emerging opportunities. The above stock selection methods are based on the analysis process of historical financial data, which makes it difficult to accurately predict industry changes or company turning points, especially in rapidly changing industries.
[0059] Therefore, although the above stock selection scheme has certain advantages in the long-term investment dimension, it lacks flexibility and real-time performance in terms of information update speed, sensitivity to market sentiment, and capturing short-term market opportunities.
[0060] The stock selection scheme based on the capital side mainly focuses on analyzing factors such as the capital flow, trading volume and capital preference of the stock market. By analyzing the above factors, the activity level of the entire market or individual stocks can be judged. In practical applications, the improvement of the capital side usually means that the market sentiment is positive and funds are willing to flow into the market, which may raise the stock price; on the contrary, the deterioration of the capital side may lead to a decline in the stock price.
[0061] However, the above stock selection scheme also has the following shortcomings:
[0062] First, the authenticity of the various factors or information on which the above scheme is based is difficult to determine. Due to the complexity of the market and the asymmetry of information, it is difficult to ensure that the financial data and materials used in the analysis of the above scheme are completely true.
[0063] Secondly, the investment strategy determined by the above scheme is not comprehensive enough. The above scheme focuses on the inflow and outflow of funds, but does not cover all factors that affect stock prices. Therefore, the investment strategy determined by the above stock selection scheme is often only auxiliary and not decisive. It is usually necessary to combine other analysis methods to improve the accuracy and reliability of the investment strategy.
[0064] Technical-based stock selection is generally achieved through data mining and visual analysis. Specifically, it mainly predicts the future performance of stocks through a variety of indicators and charts, and identifies potential stocks based on the above future performance.
[0065] The above stock selection plan is based on the assumption that market behavior will repeat, market trends are trending, and market behavior reflects all known information. However, the above three assumptions are not always true. In addition, the core of the above stock selection plan is to identify market development trends and turning points. Therefore, the execution process of the above plan usually involves the analysis of historical price and trading volume data.
[0066] The more common technical stock selection schemes include K-line analysis, moving average system, Moving Average Convergence and Divergence (MACD) indicator, and support and resistance lines, etc. The above stock selection schemes provide more objective data support for stock selection.
[0067] However, the limitations of the above stock selection scheme are also prominent, mainly reflected in the following points:
[0068] First, the above stock selection scheme has obvious limitations when facing unexpected events. For example, when the market is affected by major news or policy changes, the stock price trend may not match the analysis results of the above scheme. In this case, investors can only respond to the trend change based on their stop-loss strategy.
[0069] Secondly, the above stock selection scheme cannot provide a deterministic stock status forecast. The above stock selection scheme can only provide probabilistic information, which means that even if a certain pattern is valid in a certain historical period, it cannot be predicted that the above pattern will still recur under the same circumstances in the future period.
[0070] Thirdly, the data analysis process of the above stock selection scheme relies on historical price and trading volume data, which may result in the analysis results not accurately reflecting new information and changes in the future market. At the same time, the above stock selection scheme relies solely on technical analysis, which also leads to the neglect of other important factors such as the company's fundamental changes and macroeconomic trends.
[0071] To sum up, although the stock selection schemes in the fundamental, financial and technical dimensions provided by relevant technologies can reflect valuable market information from different angles, they all have obvious defects and limitations in practical applications.
[0072] In order to overcome the above technical problems, the relevant technology also provides a method of integrating various dimensional information through artificial intelligence (AI) models to determine the stock selection strategy. In the above scheme, it is necessary to perform multi-factor synthesis processing on various dimensional information through AI models. However, the interpretability and performance optimization of simple synthetic factor stock selection technology is still a difficult problem.
[0073] Based on the above technical problems, the embodiments of the present application provide a securities prediction method, device, equipment, medium and program product.
[0074] Figure 1 A schematic diagram of a flow chart of a securities prediction method provided in an embodiment of the present application, such as Figure 1 As shown, the method may include the following steps:
[0075] Step 101: Obtain a set of securities to be screened.
[0076] In one implementation, the type of securities included in the securities set may be greater than or equal to a first threshold; illustratively, the number of securities included in the securities set may be greater than or equal to a second threshold.
[0077] In one embodiment, the securities in a security set may be associated with the same or different industries or businesses.
[0078] In one embodiment, the securities in the securities set may include capital securities, currency securities, commodity securities, etc.; illustratively, the securities in the securities set may include stocks.
[0079] In one embodiment, the securities collection may be provided by a securities issuing institution; illustratively, the securities issuing institution may include a securities company.
[0080] Step 102: Determine screening parameters.
[0081] Among them, the screening parameters are at least associated with the institutional income of the securities issuing institution and the capital status corresponding to the securities collection; the securities issuing institution is associated with the securities collection.
[0082] In one embodiment, the institutional earnings may include the earnings of the securities issuer in at least one industry dimension within a specified time period; illustratively, the specified time period may be in monthly or annual units.
[0083] In one implementation, the fund status may include the fund flow status of the securities in the securities set; illustratively, the fund flow status may include the fund flow direction and fund flow speed, etc.
[0084] In one implementation, the securities issuing institution is associated with the securities set, which may indicate that the securities issuing institution issues the securities in the securities set, or that the securities issuing institution manages the transaction and circulation of the securities in the securities set.
[0085] In one embodiment, the securities issuers may be pre-screened and determined; illustratively, the set of securities issuers may be screened based on continuous business hours and covered industries. For example, the set of securities issuers may be screened based on the conditions that the continuous business hours are greater than half a year and the annual covered industries are greater than or equal to 20, thereby determining the securities issuers.
[0086] In one embodiment, the screening parameters can be determined by:
[0087] Determine the securities issuing institutions and industries associated with the securities collection, then obtain the institutional income of the securities issuing institutions, determine the capital flow status of the industries associated with the securities collection, perform correlation analysis on the above institutional income and capital flow status, and then determine the results of the correlation analysis as screening parameters.
[0088] Step 103: Filter the securities set based on the filtering parameters to obtain filtering results.
[0089] In one implementation, the number and / or types of securities included in the screening results may be smaller than the number and / or types of securities included in the securities set.
[0090] In one embodiment, the screening results can be obtained by any of the following methods:
[0091] Based on the first parameter in the screening parameters associated with the institutional income, an initial screening is performed on the securities collection to obtain an initial screening result, and then based on the second parameter in the screening parameters associated with the fund status, the initial screening result is screened to obtain a screening result.
[0092] Step 104: Perform prediction processing on the screening results through the prediction model to obtain a prediction result.
[0093] Among them, the yield of the securities included in the prediction results is greater than or equal to the preset threshold.
[0094] In one embodiment, the prediction model may include a neural network model with a securities return prediction function; illustratively, the neural network model may include a lightweight gradient boosting machine (Light Gradient Boosting Machine, LGBM); wherein LGBM is a framework for implementing a gradient boosting decision tree (Gradient Boosting Decision Tree, GBDT) algorithm, which supports efficient parallel training and has the advantages of faster training speed, lower memory consumption, better accuracy, and support for distribution so as to quickly process massive data.
[0095] The main idea of the GBDT algorithm is to use weak classifiers including decision trees to perform iterative training to obtain the optimal model. The algorithm has the advantages of good training effect and low overfitting. GBDT is not only widely used in the industry, but also in various data mining scenarios.
[0096] In one embodiment, the prediction result may include at least one of the type and name identification of at least one security, the industry associated with at least one security, and the rate of return of at least one security in a future period.
[0097] In one embodiment, the prediction result may include a golden stock portfolio included in a securities set; wherein the golden stock portfolio may include a recommended stock portfolio provided by a securities issuer; illustratively, the golden stock portfolio may be updated or adjusted on a monthly basis.
[0098] In one embodiment, the prediction result can be obtained by:
[0099] The historical return parameters and market situation parameters of the securities included in the screening results are obtained, and then the industry returns, the above historical return parameters and market situation parameters of the industries associated with the securities included in the screening results are analyzed and predicted through the prediction model, and the results of the analysis and prediction are determined as the prediction results.
[0100] From the above, it can be seen that the securities prediction method provided in the embodiment of the present application obtains a set of securities to be screened, and based on the screening parameters at least associated with the institutional income of the securities issuing institution and the capital status corresponding to the securities set, the securities set is screened to obtain a screening result, and the securities issuing institution is associated with the securities set. In this way, through the above operation, a double screening of the securities included in the securities set in the dimensions of securities issuing institution and capital status is achieved, thereby achieving high-precision and comprehensive screening of the securities set; and, the screening results are predicted and processed by the prediction model to obtain the prediction results, thereby reducing the amount of data processing by the prediction model in the above prediction process and improving the efficiency of the prediction model in performing prediction processing on the screening results; on the other hand, since the yield of the securities included in the prediction result is greater than or equal to the preset threshold, combined with the previous steps, the stability and accuracy of the yield of the securities included in the prediction result can be improved, thereby meeting the actual securities investment needs.
[0101] Based on the foregoing embodiments, in the securities prediction method provided in the embodiments of the present application, the screening parameters include institutional parameters associated with the institutional returns of the securities issuing institution.
[0102] In one embodiment, the institutional parameters may include dynamic fluctuation parameters of the institutional earnings over time, industry type, and market conditions.
[0103] Accordingly, determining the screening parameters can be achieved by the following steps:
[0104] Step A1: Obtain an initial securities set associated with a securities issuing institution.
[0105] In one embodiment, the type and / or quantity of securities in the initial security set may be the same as or different from the type and / or quantity of securities included in the security set.
[0106] In one embodiment, a securities issuing institution may include an institution that publishes the stability of a golden stock portfolio or golden stock data or an institution that is equal to a stability threshold; exemplarily, the number of institutions included in the securities issuing institution may be 40 to 50, and accordingly, the initial securities set may include a basic recommended stock pool provided by the securities issuing institution; exemplarily, the number of stocks included in the initial securities set may be greater than or equal to 400.
[0107] Step A2: Determine industry revenue parameters.
[0108] In one embodiment, the industry revenue parameter may correspond to a target industry; illustratively, the target industry may include at least one industry that is predetermined to have a positive impact on the society or the country's economy that is greater than or equal to a threshold.
[0109] In one embodiment, the industry revenue parameter may include a parameter of a change in the revenue of the target industry in at least one historical period.
[0110] In one embodiment, the industry revenue parameter may be determined by:
[0111] Obtain the earnings of the target industry in at least one historical period, and conduct an integrated analysis of the earnings in at least one historical period in combination with the market changes and relevant policies in the above at least one historical period, and determine the results of the integrated analysis as industry earnings parameters.
[0112] Step A3: Based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set, the industry return parameter is processed to obtain a processing result.
[0113] In one embodiment, the first industry parameter may include the industry type of the target industry corresponding to the industry revenue parameter.
[0114] In one embodiment, the second industry parameter associated with the initial securities set may include the industry type of the industry corresponding to the securities in the initial securities set.
[0115] In one embodiment, parameters in the first industry parameters may be different from at least some parameters in the second industry parameters.
[0116] In one embodiment, the processing results may include data summary results of the earnings of securities of securities issuers in at least one industry; illustratively, the above data summary results may vary or fluctuate with at least one of the type of securities, the corresponding historical period, and the market situation.
[0117] In one implementation, the industry revenue parameter is processed to obtain a processing result, which can be achieved in the following manner:
[0118] Based on the number of industry parameters that match the first industry parameter and the second industry parameter, the industry return parameter is weighted to obtain a processing result.
[0119] Step A4: Determine the mechanism parameters based on the processing results.
[0120] In one embodiment, the mechanism parameters may be determined by:
[0121] Determine the data summary results of the earnings of the securities issuer in at least one industry contained in the processing results, perform statistical averaging on the earnings data in the above data summary results in the time dimension or the industry dimension, and determine the result of the statistical averaging as the institutional parameter.
[0122] From the above, it can be seen that in the securities prediction method provided in the embodiment of the present application, after obtaining the initial securities set associated with the securities issuing institution and determining the industry return parameters, the industry return parameters are processed based on the first industry parameters associated with the industry return parameters and the second industry parameters associated with the initial securities set to obtain processing results. In this way, the flexibility and controllability of the processing of the industry return parameters can be improved; and by determining the institutional parameters based on the processing results, and the screening parameters include the institutional parameters associated with the institutional returns of the securities issuing institution, the accuracy of the institutional parameters included in the screening parameters can be improved.
[0123] Based on the foregoing embodiment, in the securities prediction method provided in the embodiment of the present application, based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set, the industry return parameter is processed to obtain a processing result, which can be achieved in the following manner:
[0124] If the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, the profit parameters in the industry profit parameters corresponding to the mth parameter in the first industry parameter are counted to obtain a first result; based on the first result, a processing result is determined.
[0125] Wherein, m and n are both integers greater than or equal to 1.
[0126] Correspondingly, if the parameters in the first industry parameters do not match the parameters in the second industry parameters, the revenue parameters corresponding to the parameters in the first industry parameters may not be counted to obtain the first result.
[0127] In one embodiment, m and n may be the same or different.
[0128] In one embodiment, the mth parameter in the first industry parameter may include the name and / or type of the industry represented by the mth data contained in the first industry parameter; accordingly, the nth parameter in the second industry parameter may represent the name and / or type of the industry represented by the nth data in the second industry parameter.
[0129] In one embodiment, the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, which can indicate that the industry name and / or type corresponding to the mth parameter in the first industry parameter is the same as the industry name and / or type corresponding to the nth parameter in the second industry parameter.
[0130] In one embodiment, the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, which may include that the industry corresponding to the initial securities set includes at least one industry in the target industry.
[0131] In one embodiment, the first result may characterize the fluctuation or change state of the revenue parameter in the industry revenue parameter corresponding to the mth parameter in the first industry parameter from at least one dimension.
[0132] In one implementation, statistics are collected on the revenue parameters in the industry revenue parameters corresponding to the mth parameter in the first industry parameter to obtain the first result, which can be achieved in the following manner:
[0133] According to the time period or industry type, the profit parameters in the industry profit parameters corresponding to the mth parameter in the first industry parameters are statistically averaged to obtain a first result; illustratively, the above time period may include monthly, quarterly or annual, etc.
[0134] In one embodiment, when the industry return parameter represents the monthly return of the stock, the first result can also be obtained in the following way:
[0135] The monthly returns of the stocks corresponding to the mth parameter in the first industry parameter are statistically averaged to obtain the first data, and then based on the difference between the first data and the index return of the industry corresponding to the mth parameter in the month, the industry excess return of the securities issuer in that month and the industry is determined, and the industry excess return of the securities issuer in that month and the industry is determined as the first result.
[0136] From the above, it can be seen that in the securities prediction method provided by the embodiment of the present application, if the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, the income parameter in the industry income parameter corresponding to the mth parameter in the first industry parameter is statistically analyzed to obtain a first result, and then the processing result is determined based on the first result. In this way, through the above operation, the controllability and screening of the income parameters in the industry income parameter can be improved, and then the accuracy of the first result and the processing result can be improved, and the flexibility of the process of obtaining the first result and determining the processing result can be improved.
[0137] Based on the foregoing embodiment, in the securities prediction method provided in the embodiment of the present application, based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set, the industry return parameter is processed to obtain a processing result, which can also be achieved in the following manner:
[0138] If the first industry parameter does not match the second industry parameter, obtain the CSI Index; determine the second result based on the CSI Index and the industry return parameters; and determine the processing result based on the second result.
[0139] Correspondingly, if the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, the processing result can be determined by the method provided in the aforementioned embodiment.
[0140] Correspondingly, if the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, the operation of obtaining the CSI Index may not be performed.
[0141] In one embodiment, if the first industry parameter does not match the second industry parameter, it can be characterized that the type of industry corresponding to the initial securities set is different from the type of industry corresponding to the target industry.
[0142] In one embodiment, the second result may represent the relative change in the index yield of the securities issuer in the industry field corresponding to the initial securities set.
[0143] In one embodiment, the second result may be determined by:
[0144] The second result is determined as the difference in index yield between the CSI All Index and the industry sector corresponding to the initial securities set.
[0145] In one embodiment, the processing result can be determined by:
[0146] The first result and the second result are statistically averaged based on the industry and the time period to obtain second data, and the second data is determined as the processing result.
[0147] In one embodiment, after determining the processing result in the above manner, the institutional parameters determined based on the processing result can be used as institutional scoring factors for associating securities issuers; in this embodiment of the present application, the securities issuers can be screened according to the above institutional scoring factors to obtain the top five issuers, and the securities data issued by the above issuers in the current month are used as the screening basis to screen the set of securities to be screened, and the golden stock pool after initial screening, that is, the initial screening result, is obtained; illustratively, the amount of stock data in the golden stock pool can be less than or equal to 100.
[0148] From the above, it can be seen that in the securities prediction method provided by the embodiment of the present application, if the first industry parameter does not match the second industry parameter, the second result is obtained and determined based on the CSI index and the industry return parameter, and the processing result is determined based on the second result. In this way, through the above steps, when the first industry parameter does not match the second industry parameter, the processing result can still be determined, thereby not only improving the stability of the process of determining the processing result, but also improving the flexibility of determining the processing result.
[0149] Based on the foregoing embodiments, in the securities prediction method provided in the embodiments of the present application, the screening parameters include capital parameters corresponding to the capital status.
[0150] In one implementation, the fund status may include a fund flow status; illustratively, the fund flow status may include at least one of a fund flow amount, a fund flow direction, and a fund flow speed.
[0151] Accordingly, the screening parameters can also be determined in the following ways:
[0152] Determine market factors associated with the state of the securities market; determine fund parameters based on the market factors.
[0153] Among them, capital parameters include stock factors or flow factors.
[0154] In one embodiment, the securities market status may include the active status of securities trading in the securities market; accordingly, the market factors may include at least one of macro factors such as the frequency of securities trading, related industry types, exchange rates, yield spreads, and treasury bond spreads.
[0155] In one embodiment, the capital parameters may include a northbound capital stock factor or a northbound capital flow factor; wherein, northbound capital refers to the sum of the Shanghai-Hong Kong Stock Connect and Shenzhen-Hong Kong Stock Connect buying funds, that is, the funds used to buy mainland stocks through the Hong Kong Stock Exchange; in actual applications, due to the regulation of the A-share market, foreign capital is not allowed to directly participate. my country opened the Shanghai-Hong Kong Stock Connect and Shenzhen-Hong Kong Stock Connect in 2014 and 2016 respectively, forming a mutual flow of funds between the A-share market and the Hong Kong stock market. Because the mainland is to the north of Hong Kong, the term northbound capital was formed.
[0156] In one embodiment, the funding parameter may be determined by:
[0157] The market situation is judged according to macro factors. At the same time, based on at least one of the macro factors in the market factors, such as the RMB exchange rate, stock market yield spread, and treasury bond spread, the dominant state of the capital stock factor or the capital flow factor is judged, and the capital stock factor or the capital flow factor is determined as the capital parameter.
[0158] Figure 2 A schematic diagram of a process for screening the initial screening results based on the capital parameters provided in the embodiment of the present application, such as Figure 2 As shown, the process may include the following steps:
[0159] Step 201, start northbound stock selection.
[0160] Exemplarily, in this step, initial screening results may be obtained.
[0161] Exemplarily, the market situation can be judged based on five macro factors including currency, credit, growth, inflation and overseas, and step 202 or 203 can be performed at one of them.
[0162] Step 202: Screening of stock factors.
[0163] Exemplarily, when the northbound capital stock factor is dominant, step 202 can be executed to screen out the top 30 stocks from the initial screening results, i.e., the screening results.
[0164] Step 203: Flow factor screening.
[0165] Exemplarily, when the northbound capital flow factor is dominant, step 203 can be executed to screen out the top 30 stocks from the initial screening results, i.e., the screening results.
[0166] Step 204: Construct a northbound gold stock portfolio.
[0167] Exemplarily, the northbound gold stock portfolio may be the prediction result in the aforementioned embodiment.
[0168] Exemplarily, the screening results can be output to the prediction model so that the prediction model can perform prediction processing on the northbound gold stock portfolio to obtain the prediction results.
[0169] Step 205: Output the northbound gold stock portfolio.
[0170] Through the above process, further screening of the initial screening results is achieved in the dimension of capital parameters.
[0171] As can be seen from the above, in the securities prediction method provided in the embodiment of the present application, after determining the market factors associated with the securities market status, the capital parameters including the stock factor or the flow factor are determined based on the market factors, and the screening parameters include the capital parameters corresponding to the capital status. In this way, through the above method, not only can the correlation between the capital parameters and the securities market status be improved, the accuracy of the capital parameters be improved, but also the comprehensiveness and accuracy of the screening parameters can be improved.
[0172] Based on the aforementioned embodiments, in the securities prediction method provided in the embodiments of the present application, the screening parameters include institutional parameters associated with the institutional income of the securities issuing institution and capital parameters corresponding to the capital status.
[0173] Accordingly, screening the securities set based on the screening parameters to obtain the screening results can be achieved in the following ways:
[0174] A first screening is performed on the securities set based on the institutional parameters to obtain a third result; and a second screening is performed on the third result based on the fund parameters to obtain a screening result.
[0175] In one implementation, the number of securities included in the third result may be smaller than the number of securities included in the securities set.
[0176] In one implementation, the third result may be obtained by:
[0177] The securities issuing institutions are screened based on the institutional returns to obtain the institutional screening results, and then a set of securities in the securities set corresponding to the institutional screening results is determined as the third result.
[0178] Exemplarily, the institution screening result may include a set of securities issuers whose returns corresponding to the target industry in the institution's returns are greater than or equal to the return threshold.
[0179] Figure 3 A schematic diagram of the process of outputting a gold stock combination provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the process may include the following steps:
[0180] Step 301: Obtain original gold stock data.
[0181] Exemplarily, the original gold stock data may be the securities set in the aforementioned embodiment.
[0182] Step 302: Screening based on institutional factors.
[0183] Exemplarily, the institutional factor may be the institutional return in the aforementioned embodiment; exemplarily, filtering the original gold stock data based on the institutional factor may obtain the third result.
[0184] Step 303: Screening based on funding factors.
[0185] Exemplarily, the funding factor may be the funding parameter in the aforementioned embodiment.
[0186] Exemplarily, the screening result may be obtained by screening the third result based on the funding factor.
[0187] Step 304: AI model processing.
[0188] Exemplarily, the AI model can be used to predict the securities included in the screening results to obtain prediction results.
[0189] Step 305: Output the gold stock combination.
[0190] Exemplarily, the golden stock combination may be the prediction result in the aforementioned embodiment.
[0191] Through the above process, double screening of the original gold stock data is achieved, which can reduce the amount of computational overhead of the AI model in processing the screening results and improve the accuracy of the AI model's prediction of the screening results.
[0192] Figure 4 A schematic diagram of a process for obtaining a gold stock combination provided in an embodiment of the present application, such as Figure 4 As shown, the process may include the following steps:
[0193] Step 401, start.
[0194] Step 402: Input screening results.
[0195] Step 403: AI model prediction.
[0196] Exemplarily, the AI model can predict the screening results to obtain predicted results.
[0197] Step 404: Sort the prediction results.
[0198] Exemplarily, the prediction result may include multiple and / or various securities. In this case, the multiple and / or various securities may be sorted based on the yields in the prediction result to obtain a sorting result.
[0199] Step 405: output the sorting result.
[0200] Exemplarily, all the sorting results may be output, or at least part of the securities in the sorting results whose yield is greater than a preset threshold may be output.
[0201] Step 406, end.
[0202] Through the above process, accurate prediction processing of the screening results is achieved, and combined with the sorting processing of the prediction results, further screening and sorting of the prediction results can be achieved.
[0203] From the above, it can be seen that the securities prediction method provided in the embodiment of the present application performs a first screening on the securities set based on the institutional parameters associated with the institutional returns of the securities issuing institution to obtain a third result, and then performs a second screening on the third result based on the capital parameters corresponding to the capital status to obtain a screening result, thereby achieving a comprehensive, diversified and progressive screening of the securities set, thereby reducing the redundancy of the data in the screening results.
[0204] Based on the foregoing embodiments, in the securities prediction method provided in the embodiments of the present application, the prediction model includes an AI model.
[0205] Accordingly, the above method may further perform the following steps:
[0206] Step B1: Obtain securities samples.
[0207] The securities sample includes multiple types of securities and securities returns associated with multiple types of securities; the number of types of securities included in the securities sample is greater than the number of types of securities included in the securities set.
[0208] In one embodiment, when the securities are stocks, the securities sample may include the full amount of daily frequency data of stocks within a specified period; illustratively, the length of the specified period may be adjusted, for example, the specified period may be 5 months.
[0209] In one embodiment, the security sample may be obtained by:
[0210] Redundant data in the full stock daily frequency data are removed, and basic features such as daily closing price, opening price, trading volume and adjusted average price in the full stock daily frequency data are retained to obtain securities samples.
[0211] Exemplarily, the securities sample may include data of CSI 300 constituent stocks between 2010-01-04 and 2018-09-28.
[0212] Step B2: Based on the training samples and verification samples in the securities samples, the initial state AI model is trained to obtain the AI model.
[0213] In one implementation, most of the samples in the securities samples may be determined as training samples, and a small number of samples in the securities samples may be determined as verification samples; illustratively, the ratio of training samples to verification samples in the securities samples may be 7:3.
[0214] In one embodiment, the process of training the initial AI model may include two steps: factor calculation and model building and tuning.
[0215]
[0216] Table 1
[0217] Among them, factor calculation can be realized by analyzing the Alpha158 factor set, and the results of factor calculation can all be technical volume and price factors; for example, the results of factor calculation can include 60-day momentum and 5-day price-volume correlation coefficient, etc. Table 1 is a list of factor sets involved in the factor calculation process.
[0218] Among them, Alpha158 is usually calculated through single-day and multi-day market data based on indicators such as opening price, closing price, highest price, lowest price, trading volume, and trading amount. It includes 158 technical quantity and price factors in multiple categories, such as single-day cross-sectional characteristics, multi-day momentum characteristics, and multi-day reversal characteristics.
[0219] Table 1 includes two columns of data: factor name and calculation formula; wherein the factor name may represent the name of the factor included in the factor set, and the calculation formula may include the formula based on which the factors in the factor set are calculated.
[0220] In one implementation, training the artificial intelligence model in the initial state may be achieved in the following manner:
[0221] Based on the securities and yields in the training sample, the model parameters of the initial state artificial intelligence model are adjusted to realize the training of the initial state artificial intelligence model and obtain preliminary training results; then, based on the securities and yields contained in the verification sample, the securities return accuracy of the preliminary training results is predicted. If the securities return accuracy meets the training objectives, the training process can be terminated and the preliminary training results are determined as the AI model; if the securities return accuracy does not meet the training objectives, the preliminary training results can continue to be trained based on the training samples.
[0222] Figure 5 A flow chart of the model training process provided in the embodiment of the present application is as follows: Figure 5 As shown, the process may include the following steps:
[0223] Step 501, start.
[0224] Step 502: Obtain initial sample data.
[0225] Exemplarily, the initial sample data may be the full amount of daily stock data within a specified period in the aforementioned embodiment.
[0226] Step 503: data preprocessing.
[0227] Exemplarily, data preprocessing can achieve the following functions:
[0228] Redundant data in the full stock daily data are eliminated, and basic features such as daily closing price, opening price, trading volume and adjusted average price in the full stock daily data are retained to obtain securities samples.
[0229] Step 504: Factor calculation.
[0230] Step 505: Model training.
[0231] Exemplarily, the AI model in the initial state can be trained based on the training samples in the securities sample to perform the model training process.
[0232] Step 506: Input verification sample.
[0233] Step 507: Verify the model and adjust parameters.
[0234] Step 508: Save the parameters.
[0235] Exemplarily, the above parameters may include model parameters that ensure that the yield prediction results of the AI model during the training process meet the training requirements.
[0236] Step 509: Obtain an AI model.
[0237] Step 510, end.
[0238] Through the above process, the integrated execution of AI model training and verification is achieved, thereby improving the prediction accuracy of the AI model.
[0239] From the above, it can be seen that the securities prediction method provided in the embodiment of the present application obtains securities samples including multiple types of securities and securities returns associated with multiple types of securities, and then trains the initial state AI model based on the training samples and verification samples in the securities samples to obtain the AI model, and the number of types of securities included in the securities samples. In this way, not only can the richness of the securities samples be improved, but also, by training the initial state AI model based on securities samples with higher richness, the accuracy of the AI model obtained after the training in securities prediction can be improved.
[0240] Figure 6Schematic diagram of the comparison effect of the yield of securities determined by the securities prediction method provided by the embodiment of the present application, such as Figure 6 As shown in Figure 2, the period corresponding to this rate of return can cover the period from January 30, 2018 to September 30, 2022; Figure 6 In the figure, the first curve 601 is used to characterize the yield of the securities in the above period determined by the securities prediction method provided by the embodiment of the present application, the second curve 602 is used to characterize the excess yield, and the third curve 603 is used to characterize the yield of the CSI 300.
[0241] From the comparison between the first curve 601 and the third curve 603, it can be seen that the yield of the securities determined by the securities prediction method provided in the embodiment of the present application during the above period is higher than the excess yield and the CSI 300 yield.
[0242] Table 2 is the statistical data results of the securities yield determined by the securities prediction method provided by the embodiment of the present application.
[0243] Table 3 is the statistical results of the securities yields between 2019 and 2022 determined by the securities forecasting method provided in the embodiment of the present application.
[0244] Total yield 252% Annualized Rate of Return 35% Annual Volatility 25% Win Rate 56% Profit and loss ratio 1.12 Maximum Drawdown Rate 35% Sharpe Ratio 1.37 Calmar ratio 1
[0245] Table 2
[0246] years Yield Public Fund Ranking 2019 69.49% 44 / 326 2020 59.36% 197 / 390 2021 64.61% 5 / 534 2022 -13.25 232 / 858
[0247] Table 3
[0248] It can be seen from Table 3 that before 2019 to 2022, in most years, the yields of securities determined by the securities prediction method provided in the embodiments of the present application were relatively high.
[0249] Figure 7 FIG. 2 is another schematic diagram showing the comparison effect of the yield of securities determined by the securities prediction method provided by the embodiment of the present application. Figure 7 As shown, the fourth curve 701 represents the yield of the securities determined by the securities prediction method provided in the embodiment of the present application between May 9, 2023 and May 9, 2024, and the fifth curve 702 represents the yield of the CSI All-Share Index between May 9, 2023 and May 9, 2024. Figure 7 It can be seen that the annual yield from May 9, 2023 to May 9, 2024, relative to the yield, winning rate and excess yield of the CSI All Index in the above years, the annual yield, winning rate and excess yield of the securities determined by the securities prediction method provided in the embodiment of the present application are 11.54%, 55.83% and 22.09% respectively.
[0250] Figure 8 This is a schematic diagram of the monthly revenue comparison data summary effect provided in the embodiment of the present application. Figure 8 The months from 202306 to 202405 are listed and explained from the dimensions of monthly, winning rate, average maximum increase of individual stocks, strategy return, benchmark return (CSI All Index), excess return, maximum strategy return and maximum excess return; among them, the strategy return can represent the yield of the securities determined by the securities prediction method provided in the embodiment of the present application, and the maximum strategy return can represent the maximum yield determined by the method provided in the embodiment of the present application in a certain month.
[0251] from Figure 8 It can be seen that in Figure 8 In each of the months shown, the winning rates of the securities determined by the method provided in this application are all high, and the strategy returns show positive returns in most of the months, and the maximum strategy returns show a positive return trend in the vast majority of the months.
[0252] From the above, it can be seen that the rate of return and winning rate of securities determined by the securities prediction method provided in the embodiment of the present application can realize the accurate screening of massive securities, thereby meeting the actual investment needs of securities investment users.
[0253] Based on the above embodiments, the present application also provides a securities prediction device. Fig. 9 A schematic diagram of the structure of the securities prediction device provided in the embodiment of the present application is shown in FIG. Fig. 9 As shown, the securities prediction device 9 may include:
[0254] An acquisition module 901 is used to acquire a set of securities to be screened;
[0255] Determination module 902, for determining screening parameters; screening the securities set based on the screening parameters to obtain screening results; wherein the screening parameters are at least associated with the institutional income of the securities issuing institution and the capital status corresponding to the securities set; the securities issuing institution is associated with the securities set;
[0256] The prediction module 903 is used to perform prediction processing on the screening results through the prediction model to obtain the prediction results; wherein the yield of the securities included in the prediction results is greater than or equal to a preset threshold.
[0257] In some embodiments, the screening parameters include an institutional parameter associated with an institutional return of the security issuer;
[0258] An acquisition module 901 is used to acquire an initial securities set associated with a securities issuing institution;
[0259] Determination module 902 is used to determine industry return parameters; based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set, the industry return parameter is processed to obtain a processing result; based on the processing result, the institutional parameter is determined.
[0260] In some embodiments, the determination module 902 is used to, if the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, perform statistics on the revenue parameter in the industry revenue parameter corresponding to the mth parameter in the first industry parameter to obtain a first result; wherein m and n are both integers greater than or equal to 1;
[0261] The determination module 902 is further configured to determine a processing result based on the first result.
[0262] In some embodiments, the acquisition module 901 is used to acquire the CSI All-Share Index if the first industry parameter does not match the second industry parameter;
[0263] Determination module 902 is used to determine the second result based on the CSI All-Share Index and industry return parameters; and determine the processing result based on the second result.
[0264] In some embodiments, the determination module 902 is used to determine market factors associated with the securities market status; determine capital parameters based on the market factors; wherein the capital parameters include stock factors or flow factors.
[0265] In some embodiments, the screening parameters include an institutional parameter associated with an institutional return of the securities issuer and a funding parameter corresponding to a funding status;
[0266] The determination module 902 is used to perform a first screening on the securities set based on the institutional parameters to obtain a third result; and perform a second screening on the third result based on the fund parameters to obtain a screening result.
[0267] In some embodiments, the predictive model comprises an artificial intelligence model;
[0268] The acquisition module 901 is used to acquire a securities sample; wherein the securities sample includes multiple types of securities and securities returns associated with the multiple types of securities; the number of types of securities included in the securities sample is greater than the number of types of securities included in the securities set;
[0269] The determination module 902 is used to train the artificial intelligence model in the initial state based on the training samples and the verification samples in the securities samples to obtain the artificial intelligence model.
[0270] Based on the above embodiments, the present application also provides an electronic device, Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Fig.10As shown, the electronic device 10 may include a processor 1001 and a memory 1002; wherein the memory 1002 stores a computer program; when the computer program is executed by the processor 1001, it can implement the securities prediction method provided in any of the previous embodiments.
[0271] Based on the foregoing embodiments, the embodiments of the present application further provide a computer-readable storage medium; a computer program is stored in the storage medium; when the computer program is executed by a processor of an electronic device, it can implement a securities prediction method as provided in any of the previous embodiments.
[0272] Based on the foregoing embodiments, the embodiments of the present application also provide a computer program product; the program product includes a computer program; when the computer program is executed by a processor of an electronic device, it can implement the securities prediction method provided in any of the previous embodiments.
[0273] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.
[0274] The methods disclosed in the various method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.
[0275] The features disclosed in the various product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.
[0276] The features disclosed in the various method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.
[0277] It should be noted that the above-mentioned computer-readable storage medium can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM) and other memories; it can also be various electronic devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.
[0278] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0279] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0280] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus necessary general hardware nodes, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.
[0281] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0282] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0283] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0284] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A securities prediction method, characterized in that: The method comprises: Get the securities collection to be screened; Determine a screening parameter; wherein the screening parameter is at least associated with the institutional income of the securities issuing institution and the fund status corresponding to the securities set; the securities issuing institution is associated with the securities set; Screening the securities set based on the screening parameters to obtain screening results; The screening results are predicted and processed by a prediction model to obtain a prediction result; wherein the yield of the securities included in the prediction result is greater than or equal to a preset threshold.
2. The method according to claim 1, characterized in that: The screening parameters include an institutional parameter associated with the institutional income of the securities issuing institution; the determining of the screening parameters includes: Obtaining an initial securities set associated with the securities issuing institution; Determine industry revenue parameters; Based on the first industry parameter associated with the industry return parameter and the second industry parameter associated with the initial securities set, the industry return parameter is processed to obtain a processing result; Based on the processing results, the mechanism parameters are determined.
3. The method according to claim 2, characterized in that The processing of the industry income parameter based on the first industry parameter associated with the industry income parameter and the second industry parameter associated with the initial securities set to obtain a processing result includes: If the mth parameter in the first industry parameter matches the nth parameter in the second industry parameter, the revenue parameter in the industry revenue parameter corresponding to the mth parameter in the first industry parameter is counted to obtain a first result; wherein m and n are both integers greater than or equal to 1; Based on the first result, the processing result is determined.
4. The method according to claim 2, characterized in that The processing of the industry income parameter based on the first industry parameter associated with the industry income parameter and the second industry parameter associated with the initial securities set to obtain a processing result includes: If the first industry parameter does not match the second industry parameter, obtain the CSI All-Share Index; Determine a second result based on the CSI All-Share Index and the industry return parameter; Based on the second result, the processing result is determined.
5. The method according to claim 1, characterized in that The screening parameters include the fund parameters corresponding to the fund status; the determining of the screening parameters includes: Identify market factors associated with security market conditions; The funding parameter is determined based on the market factor; wherein the funding parameter includes a stock factor or a flow factor.
6. The method according to claim 1, characterized in that The screening parameters include an institutional parameter associated with the institutional income of the securities issuing institution and a capital parameter corresponding to the capital status; the securities set is screened based on the screening parameters to obtain a screening result, including: Performing a first screening on the securities set based on the institutional parameter to obtain a third result; A second screening is performed on the third result based on the fund parameter to obtain the screening result.
7. The method according to claim 1, characterized in that The prediction model includes an artificial intelligence model; the method further includes: Acquire a securities sample; wherein the securities sample includes multiple types of securities and securities returns associated with the multiple types of securities; the number of types of securities included in the securities sample is greater than the number of types of securities included in the securities set; Based on the training samples and verification samples in the securities sample, the artificial intelligence model in the initial state is trained to obtain the artificial intelligence model.
8. A securities prediction device, characterized in that: The securities prediction device comprises: An acquisition module is used to acquire a set of securities to be screened; A determination module, used to determine a screening parameter; based on the screening parameter, the securities set is screened to obtain a screening result; wherein the screening parameter is at least associated with the institutional income of the securities issuing institution and the capital status corresponding to the securities set; the securities issuing institution is associated with the securities set; A prediction module is used to perform prediction processing on the screening results through a prediction model to obtain a prediction result; wherein the yield of the securities included in the prediction result is greater than or equal to a preset threshold.
9. An electronic device, characterized in that: The electronic device includes a processor and a memory; a computer program is stored in the memory; when the computer program is executed by the processor, it can implement the securities prediction method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program; when the computer program is executed by a processor of an electronic device, it can implement the securities prediction method as described in any one of claims 1 to 7.
11. A computer program product, characterized in that The computer program product includes a computer program; when the computer program is executed by a processor of an electronic device, it can implement the securities prediction method as described in any one of claims 1 to 7.
