Value investment and transaction decision-making method and system based on behavior finance

Abstract

The invention relates to a value investment and transaction decision-making method and system based on behavioral finance, relates to the field of investment management, and solves the problems of low data processing and analysis efficiency and lack of scientificity of decision-making in related technologies. Comprising the following steps: in response to a preset transaction triggering condition, obtaining financial data of each investment target starting the triggering condition, and determining a market performance score of the investment target according to the financial data of the investment target and a preset weight; determining a quality factor QMJ index according to the financial data of the investment target; determining internal value data of the investment target according to the financial data of the investment target by using a cash flow discounting model; and determining a transaction decision of the investment target according to the financial data, the market performance score, the quality factor QMJ index and the intrinsic value data of the investment target. According to the technical scheme, systematization, scientization and rigorization of investment research and transaction decision making are achieved, the accuracy and efficiency of investment decision making are improved, and the investment risk is reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of investment management, and in particular to a value investment and trading decision-making method and system based on behavioral finance. Background Art

[0002] In the field of financial investment research and fund management, traditional methods have many shortcomings.

[0003] 1. The lack of a systematic process leads to irregularities in the investment decision-making process, making it difficult to ensure scientificity and rigor.

[0004] 2. The processing and analysis of large amounts of financial data are not efficient enough, and cannot provide timely and accurate decision support to investors.

[0005] 3. In terms of stock research, there is a lack of comprehensive analysis and tracking of various stock indicators, making it difficult for users and analysts to quickly obtain comprehensive stock information.

[0006] 4. In fund transaction management, transaction decisions lack scientific basis and transaction processes are not standardized, which easily leads to investment risks. Summary of the Invention

[0007] In order to overcome the problems existing in the related technologies, the present disclosure provides a value investment and trading decision-making method and system based on behavioral finance to solve the problems of low efficiency in data processing and analysis and lack of scientific decision-making in the related technologies.

[0008] According to a first aspect of an embodiment of the present disclosure, a value investment and trading decision-making method based on behavioral finance is provided, comprising:

[0009] In response to a preset transaction trigger condition, obtaining financial data of each investment target that initiated the trigger condition, the financial data including: basic information, financial indicator data, volume and price indicator data, and shareholder transaction data;

[0010] Determine the market performance score of the investment target based on the financial indicator data and the quantity and price indicator data in the financial data of the investment target and the preset weights;

[0011] Determine the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target;

[0012] Determining intrinsic value data of the investment target based on the financial data of the investment target using a discounted cash flow model, wherein the intrinsic value data indicates whether the investment value of the investment target is underestimated or overestimated;

[0013] The trading decision of the investment target is determined based on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target. The trading decision may include one or more of the following: whether to buy, whether to sell, whether to hold a position, the timing of buying or selling, and the quantity and price of buying or selling.

[0014] Preferably, the transaction triggering condition includes at least one of the following:

[0015] The scheduled time or period arrives;

[0016] The change in the quality of the individual stocks of the investment target meets the first condition;

[0017] The change in the valuation of the investment target satisfies the second condition;

[0018] The change in the behavioral signal of the investment target satisfies the third condition;

[0019] The expected return change of the investment target satisfies the fourth condition;

[0020] The basic information includes at least one of the following:

[0021] Code, name, sector, listing date, and industry classification;

[0022] The financial indicator data includes at least one of the following:

[0023] Profit before interest and taxes, operating income, net profit, return on equity (ROE), return on total assets (ROA), gross profit margin;

[0024] The quantity and price indicator data includes at least one of the following:

[0025] Trading volume, closing price, opening price, highest price, lowest price;

[0026] The shareholder transaction data includes: the cumulative increase or decrease in shareholdings by shareholders and executives within a predetermined period of time.

[0027] Preferably, according to the financial indicator data and the quantity and price indicator data in the financial data of the investment target,

[0028] and preset weights to determine the market performance score of the investment target, including:

[0029] According to the rise and fall of each indicator data in the financial indicator data and the quantity and price indicator data, numerical statistics are performed, wherein the rise is recorded as the first value and the fall is recorded as the second value;

[0030] The market performance score of the investment target is calculated by taking the weighted sum of the numerical value of each indicator data and the weight of each indicator number.

[0031] Preferably, determining the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target includes:

[0032] Calculate the rolling 12-month gross profit per unit of total assets by dividing the gross profit and total assets data in the financial indicator data;

[0033] The growth rate of ROE for rolling 12 months is calculated by comparing ROE of different time periods;

[0034] The volatility of the ROE for the rolling 12 months is calculated by comparing the ROE of different time periods.

[0035] Preferably, using a discounted cash flow model, based on the financial data of the investment target, the intrinsic value data of the investment target is determined, wherein the intrinsic value data reflects whether the investment value of the investment target is underestimated or overestimated, including:

[0036] By inputting the predicted future free cash flow FCFF, the predicted future weighted average cost of capital WACC, the year-on-year growth rate of operating income, the profit margin before interest and taxes, and the operating income / invested capital less cash and equivalents, and using the cash flow discount model, the intrinsic value data Z of the investment target is calculated according to the following formula:

[0037] Z=∑[FCFF(t) / (1+WACC)^t]+final value / (1+WACC)^n;

[0038] Where t represents the year of the forecast period and n represents the last year of the forecast period.

[0039] Preferably, determining a trading decision for an investment target based on the financial data, market performance score, quality factor QMJ indicator, and intrinsic value data of the investment target includes:

[0040] Input the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target into the preset dynamic memory-focus synergy model to determine the research coefficient of the decision-making model;

[0041] Determine trading decisions based on investment objectives based on subscription and redemption setting information and trading decision conditions.

[0042] Preferably, the decision model is a buyer's extraordinary research model or a seller's extraordinary research model;

[0043] The financial data, market performance scores, quality factor QMJ indicators and intrinsic value data of the investment targets are normalized using a rolling window, with the window length W aligned with the financial reporting period;

[0044] Perform winsoring on extreme values ​​in the financial data, market performance score, quality factor QMJ indicator, and intrinsic value data of the investment target, using a preset quantile cutoff;

[0045] A dynamic memory gated temporal encoder with macroeconomic gating is used to perform nonlinear coupling analysis on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target;

[0046] The nonlinear coupling analysis results are processed in time channel and cross-sectional channel by a dual-channel adaptive feature focuser to generate the survey coefficients of the decision model, where:

[0047] Time channel: Analyze the contribution of the investment target's financial data, market performance score, quality factor QMJ indicator and intrinsic value data in the economic cycle;

[0048] Cross-sectional channel: Calculate the relative ranking of the investment target's financial data, market performance score, quality factor QMJ indicator and intrinsic value data within the industry.

[0049] Preferably, the buyer's extraordinary research model is expressed as:

[0050] LNMEETING i,m

[0051] =β0+β1SIZE i,m +β2TURN i,m +β3MOMEN i,m +β4ROA i,m

[0052] +∈ i,m (AIM)

[0053] Among them, LNMEETING i,m It represents the logarithm of the number of all investor surveys in the past 90 days for the i-th investment target at the end of the m-th month;

[0054] SIZE i,m : represents the logarithm of the circulating market value of the i-th investment target at the end of the m-th month;

[0055] TURN i,m : represents the average turnover rate of the i-th investment target at the end of the m-th month in the past 12 months;

[0056] MOMEN i,m : represents the cumulative stock return rate of the i-th investment target at the end of the m-th month, over a rolling 12-month period;

[0057] ROA i,m: represents the return on total assets (ROA) of the i-th investment target at the end of month m over a rolling 12-month period;

[0058] ∈ i,m (AIM): represents the value of the buy-side extraordinary survey of the i-th investment target at the end of month m;

[0059] β0, β1, β2, β3, and β4 are the buyer’s extraordinary research coefficients;

[0060] When ∈ i,m (AIM) is greater than the set buying threshold, or ∈ i,m When the ranking of (AIM) is higher than the preset ranking of all investment targets, confirm the purchase;

[0061] When ∈ i,m (AIM) is less than the set sell threshold, or ∈ i,m When the ranking of (AIM) is below the preset ranking of all investment targets, it is determined to sell;

[0062] The buyer's extraordinary research model is expressed as:

[0063] Log(1+TOT i,m

[0064] =α0+α1SIZE i,m +α2TURN i,m +α3MOMEN i,m

[0065] +∈ i,m (ATOT)

[0066] Among them, TOT i,m : represents the number of analysts covering the i-th investment target in the past 90 days at the end of the m-th month;

[0067] ∈ i,m (ATOT) represents the value of the sell-side extraordinary survey of the i-th investment target at the end of month m;

[0068] α0, α1, α2, and α3 are the seller’s extraordinary research coefficients;

[0069] When ∈ i,m When (ATOT) is greater than the set buying threshold, or ∈ i,m When the ranking of (AIM) is higher than the preset ranking of all investment targets, confirm the purchase;

[0070] When ∈ i,m When (ATOT) is less than the set selling threshold, or ∈ i,m When the ranking of (AIM) is below the preset ranking of all investment targets, it is determined to sell.

[0071] Preferably, the value investment and trading decision-making method based on behavioral finance is characterized by further comprising: obtaining investment research related information; the investment research related information at least includes: company action event information;

[0072] When determining the trading decision for an investment target, a trading decision analysis is conducted based on the company's action event information, the strategic decision and operational dynamics of the investment target, and the investment target.

[0073] In a second aspect, the present invention further provides a value investment and trading decision-making system based on behavioral finance, comprising:

[0074] Investment research task management, which is used to obtain financial data of each investment target that triggers the trigger condition in response to the preset transaction trigger condition. The financial data includes: basic information, financial indicator data, volume and price indicator data, and shareholder transaction data;

[0075] An individual stock research module is used to determine the market performance score of an investment target based on the financial indicator data and the volume and price indicator data in the financial data of the investment target and the preset weights;

[0076] The individual stock research module is also used to determine the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target;

[0077] The individual stock research module is further configured to determine intrinsic value data of an investment target based on the financial data of the investment target using a discounted cash flow model, wherein the intrinsic value data indicates whether the investment value of the investment target is underestimated or overestimated;

[0078] A transaction management module is used to determine transaction decisions for investment targets based on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target. The transaction decisions may include one or more of the following: whether to buy, whether to sell, whether to hold a position, the timing of buying or selling, and the quantity and price of buying or selling.

[0079] The technical solutions provided by the embodiments of the present disclosure may have the following beneficial effects:

[0080] The above-mentioned technical solution of the present application can solve the problems existing in the process of financial investment research and fund management, such as non-standardized processes, low efficiency of data processing and analysis, lack of scientific decision-making, and insufficient functional integration. By constructing a technical solution that integrates investment research and trading decisions, it can achieve systematization, scientificization, intelligence and rigor of investment research and trading decisions, improve the accuracy and efficiency of investment decisions, and reduce investment risks. With the help of behavioral signals in behavioral finance, the accuracy of investment research is increased. Through the data processing and analysis process of the financial data of the investment target, in order to help investors make more accurate and efficient value investments, this application is based on the theory of behavioral finance and uses intelligent technology as a tool to help investors better conduct value investments and provide strong support for investment decisions. In addition, the technical solution of the present invention can modularly participate in various links of the transaction management of investment targets to meet the diverse needs of different investors.

[0081] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0083] Figure 1 is a flow chart showing a value investment and trading decision-making method based on behavioral finance according to an exemplary embodiment;

[0084] Figure 2 The figure is a schematic diagram of a value investment and trading decision-making system based on behavioral finance according to an exemplary embodiment. DETAILED DESCRIPTION

[0085] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

[0086] See attached Figure 1 , provides a value investment and trading decision-making method based on behavioral finance, which can include the following steps:

[0087] S101. In response to a preset transaction trigger condition, obtain financial data of each investment target that triggers the trigger condition, wherein the financial data includes basic information, financial indicator data, volume and price indicator data, and shareholder transaction data;

[0088] S102: Determine a market performance score of the investment target based on the financial indicator data and the quantity and price indicator data in the financial data of the investment target and preset weights;

[0089] S103, determining a quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target;

[0090] S104. Determine intrinsic value data of the investment target based on the financial data of the investment target using a discounted cash flow model, wherein the intrinsic value data indicates whether the investment value of the investment target is underestimated or overestimated;

[0091] S105. Determine a trading decision for the investment target based on the financial data, market performance score, quality factor QMJ indicator, and intrinsic value data of the investment target. The trading decision may include one or more of the following: whether to buy, whether to sell, whether to hold a position, the timing of buying or selling, and the quantity and price of buying or selling.

[0092] In the field of financial investment research and fund management, some related technologies focus on data analysis in investment research while neglecting fund transaction management. Alternatively, the setting of transaction trigger conditions in fund transaction management is not flexible enough, failing to meet the diverse needs of different investors. Consequently, deep integration of investment research and fund management cannot be achieved, making it difficult to form complete, scientific, and rigorous investment decisions. The technical solutions of the embodiments of the present invention can address issues such as irregular processes, inefficient data processing and analysis, unscientific decision-making, and insufficient functional integration in the financial investment research and fund management processes. By constructing a technical solution that integrates investment research and transaction decision-making, investment research and transaction decision-making can be systematic, scientific, intelligent, and rigorous, improving the accuracy and efficiency of investment decisions and reducing investment risks. By leveraging behavioral signals from behavioral finance, the accuracy of investment research can be increased. By processing and analyzing the financial data of investment targets, and in order to help investors make more accurate and efficient value investments, this application, based on the theory of behavioral finance and using intelligent technology as a tool, helps investors better conduct value investing and provides strong support for investment decisions. Furthermore, the technical solution of the present invention can modularize each link of the transaction management of investment targets to meet the diverse needs of different investors.

[0093] In an embodiment of the present invention, the transaction triggering condition includes at least one of the following:

[0094] The scheduled time or period arrives;

[0095] The change in the quality of the individual stocks of the investment target meets the first condition;

[0096] The change in the valuation of the investment target satisfies the second condition;

[0097] The change in the behavioral signal of the investment target satisfies the third condition;

[0098] The expected return change of the investment target satisfies the fourth condition;

[0099] The basic information includes at least one of the following:

[0100] Code, name, sector, listing date, and industry classification;

[0101] The financial indicator data includes at least one of the following:

[0102] Profit before interest and taxes, operating income, net profit, return on equity (ROE), return on total assets (ROA), gross profit margin;

[0103] The quantity and price indicator data includes at least one of the following:

[0104] Trading volume, closing price, opening price, highest price, lowest price;

[0105] The shareholder transaction data includes: the cumulative increase or decrease in the proportion of shares held by shareholders and senior executives within a predetermined period of time. In an embodiment of the present invention, changes in the quality of individual stocks of investment targets are manifested as: 1. the occurrence of corporate action events; 2. changes in fundamentals. Changes in the valuation of investment targets are manifested as: changes in valuation. Changes in behavioral signals of investment targets are manifested as: 1. increases or decreases in holdings, share prices and share buyback ratios of important shareholders and senior executives of the company; 2. changes in the increase, decrease and continuous holdings of major institutions; 3. changes in the consensus forecasts of sell-side analysts and the extraordinary coverage values ​​of sell-side analysts; 4. changes in the volume and price indicators of individual stocks in the market. Changes in the expected returns of investment targets are manifested as: 1. changes in the safety margin; 2. changes in the expected future rate of return; 3. changes in the implied growth rate of the stock price. In an embodiment of the present invention, the parameters or proportions of each event can be set so that when the conditions are met, the acquisition of financial data and subsequent steps are triggered.

[0106] In this embodiment of the present invention, the financial data in step S101 is sourced from an authoritative financial data platform to ensure data accuracy and timeliness. This embodiment of the present invention transmits data through an interface with the data platform, acquiring the latest data in real time. For example, it can obtain basic information, financial indicators, volume and price indicators, and shareholder transaction data for all stocks over the past three years, including but not limited to market data, economic database (EDB) data, and financial data for each stock or investment target. Investment targets can be determined based on user-defined stock pools in the system, such as those designated as already invested, focused, or of excellent quality, or from authoritative rankings. Users can also manually enter a stock code or name to add a new one, or all currently available stocks can be selected. The financial data acquired by this embodiment of the present invention is stored in a database or server. The database or server utilizes an efficient storage structure to categorize and store each stock's native indicators, including market data such as stock code, stock name, timestamp, closing price, and trading volume, as well as financial indicators such as total tonnage margin (TTM), consensus forecasted operating income, and two-year compound annual growth rate, to facilitate subsequent data query and processing. It also has a data verification function, which allows you to query the values ​​of native indicators and calculated indicator data in the database by entering stock codes and timestamps.

[0107] In this embodiment of the present invention, using stocks as an example, basic information about investment targets includes stock code, stock name, sector, listing date, and industry classification. This basic information can be viewed by users, for example, by presenting data within the basic information in an interactive interface, allowing for a preliminary understanding and classification of the stock. Financial indicator data includes earnings before interest and taxes, operating income, net profit, return on equity (ROE), return on assets (ROA), and gross profit margin. This data can also be presented in an interactive interface, such as as data or charts, to assess the stock's profitability and financial status. Volume and price indicator data includes trading volume, closing price, opening price, highest price, and lowest price. This data is used in the subsequent calculation of indicators and data such as market performance scores to analyze market trends and stock price movements. Shareholder transaction data includes the cumulative increase or decrease in shareholdings by shareholders and executives over a predetermined period, such as the cumulative increase or decrease in shareholdings by all major shareholders and executives over the past 1, 3, 6, 12, or 18 months. This data can also be presented in an interactive interface to analyze insider trading of stocks and assist in determining their investment value.

[0108] The transaction trigger condition in the embodiment of the present invention can be the arrival of a predetermined time or cycle, for example: stock market data is obtained regularly (such as before the opening or after the closing of the market every day) through the interface with the financial data platform. First, the transaction trigger condition is initiated (the cycle or predetermined time is reached) and a request is sent to the financial data platform. The request contains parameters such as the time range for obtaining data (the same day, a year or a custom time period), the stock range (all stocks or target stocks), etc. After receiving the request, the financial data platform filters and returns the corresponding data according to the parameters. After receiving the data, the data is preliminarily verified and cleaned to ensure the integrity and accuracy of the data. The verified and cleaned data is stored in the database according to the preset storage structure. For the relational database part, the basic information of the stock, financial indicator data and other structured data are stored in the corresponding table, and the stock code is used as the primary key for association. For unstructured data, such as research reports and profit forecast files uploaded by users, they are stored in the distributed file system, and the storage path and related metadata of the file are recorded in the relational database.

[0109] Specifically, taking stocks as an example, basic stock information is read from the database, extracting data such as the stock code, stock name, and sector, and storing it in a pre-set data structure. This facilitates subsequent module calls, allowing these data to be retrieved directly from memory for display. Financial indicator data is extracted from the database using SQL queries. For example, data such as EBITDA and operating income is queried and extracted using SQL SELECT statements based on the table structure of the financial statement data storage. The extracted data is used in page and chart calculations. Volume and price indicator data is also extracted using SQL queries. Based on the structure of the market data table, data such as trading volume and closing price are filtered by timestamp and stock code. This extracted data is used to calculate the market index, volume, and price indicators in the market overview module. Shareholder and executive transaction data is stored in a dedicated database table. Queries are written to extract data on shareholding increases and decreases based on time ranges and stock codes. The extracted data is used in page and chart calculations, querying the database in real time to obtain the latest data.

[0110] In an embodiment of the present invention, determining the market performance score of an investment target based on the financial indicator data and the quantity and price indicator data in the financial data of the investment target and preset weights includes:

[0111] According to the rise and fall of each indicator data in the financial indicator data and the quantity and price indicator data, numerical statistics are performed, wherein the rise is recorded as the first value and the fall is recorded as the second value;

[0112] The market performance score of the investment target is calculated by taking the weighted sum of the numerical value of each indicator data and the weight of each indicator number.

[0113] Based on the market index, volume, and price indicators and preset weights (e.g., a weight of 0.30 for the moving average (MA), 0.20 for the rate of change (ROC), 0.20 for the moving average convergence / divergence (MACD), 0.10 for the bias (BIAS), 0.10 for the triple exponential moving average (TRIX), and 0.10 for the momentum indicator (MTM), the market performance score is calculated by multiplying the score by a coefficient, assigning 1 point for an increase and 0 point for a decrease. This score serves as a key indicator for evaluating stock market performance. For example, if a stock's moving average (MA) indicates an increase, its rate of change (ROC) indicates a decrease, and its MACD indicates an increase over a certain period of time, the market performance score is calculated based on these rules and weights, thereby evaluating its market performance.

[0114] Specifically, the system first obtains the market index's volume and price indicators and preset weighting data. Then, it assigns a score based on the index's rise or fall (1 point for an increase, 0 point for a decrease). This score is then multiplied by the corresponding weighting, and finally, the scores for each indicator are summed to obtain the stock market performance score. This calculation process can be implemented by writing a function in a programming language such as Python or Java, which takes the volume and price indicator data and weighting data as input and outputs the calculated stock market performance score.

[0115] In an embodiment of the present invention, determining the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target includes:

[0116] Calculate the rolling 12-month gross profit per unit of total assets by dividing the gross profit and total assets data in the financial indicator data;

[0117] The growth rate of ROE for rolling 12 months is calculated by comparing ROE of different time periods;

[0118] The volatility of the ROE for the rolling 12 months is calculated by comparing the ROE of different time periods.

[0119] In this embodiment of the present invention, various quality factors (QMJ) are generated by analyzing and calculating extracted financial indicator data. For example, indicators such as gross profit per unit (TTM), return on equity (TTM) growth rate, and ROE volatility (over the past three years) are calculated. These indicators reflect the quality, growth potential, and profitability of a stock from different perspectives. The calculation and analysis of these indicators provides a deeper understanding of a stock's intrinsic value and development trends.

[0120] In the embodiment of the present invention, the QMJ security score ranking is calculated; the five sub-items of 1, "Negative Stock Price Beta (minus_beta)", 2, "Negative Total Debt to Total Assets Ratio (MRQ) (minus_tot_liab_over_tot_assets)", 3, "Negative M Score (MRQ) (minus_m_score)", 4, "Z Value (MRQ) (z_score)", and 5, "Negative ROE Volatility (Last 3 Years) (minus_qfa_roe_deducted_stdev)" are ranked from large to small (from high to low), and then the ranking is Z-transformed to obtain the Z score; the scores (z scores) of each sub-item are then summed up and the arithmetic average is calculated to obtain the total QMJ security score (z score) of each listed company in each period.

[0121] In embodiments of the present invention, Python data analysis libraries (such as Pandas and NumPy) can be used to calculate the extracted financial indicator data. For example, when calculating the gross profit per unit (TTM), a division operation is performed based on the gross profit and total assets data in the financial indicator data; when calculating the growth rate of return on equity (TTM), the return on equity data is calculated by comparing the return on equity data of different time periods. After the calculation is completed, this data is used to generate the QMJ indicator chart.

[0122] In an embodiment of the present invention, intrinsic value data of an investment target is determined based on the financial data of the investment target using a discounted cash flow model. The intrinsic value data reflects whether the investment value of the investment target is underestimated or overestimated, including:

[0123] By inputting data such as the predicted future free cash flow (FCFF), the predicted future weighted average cost of capital (WACC), the year-on-year growth rate of operating income, the EBIT margin, and the ratio of operating income to invested capital less cash and equivalents (part of which is derived from extracted financial data and part of which is user input), and using the discounted cash flow DCF model, the intrinsic value data Z of the investment target is calculated according to the following formula:

[0124] Z=∑[FCFF(t) / (1+WACC)^t]+final value / (1+WACC)^n;

[0125] Where t represents the year of the forecast period and n represents the last year of the forecast period.

[0126] In this embodiment of the present invention, the DCF model is used to calculate data such as the company's intrinsic value by forecasting future free cash flow (FCFF), future weighted average cost of capital (WACC), year-on-year operating income growth rate, EBITDA margin, and operating income / invested capital less cash and equivalents. This data provides an important basis for stock valuation, helping investors determine whether a stock's investment value is undervalued or overvalued.

[0127] In the embodiment of the present invention, after obtaining the data such as the predicted future free cash flow (FCFF) and the predicted future weighted average cost of capital (WACC), the calculation formula of the DCF model is used to calculate the intrinsic value data of the stock using an algorithm written in Python or other programming languages. For example, the intrinsic value data of the stock is calculated according to the formula:

[0128] Z=∑[FCFF(t) / (1+WACC)^t]+final value / (1+WACC)^n,

[0129] Where t represents the year of the forecast period, and n represents the final year of the forecast period. The calculated intrinsic value data, Z, is displayed on the page to provide users with a stock valuation reference.

[0130] Forecasted future corporate free cash flow (FCFF) refers to the remaining cash flow from a company's operating activities available to all investors (including shareholders and creditors) after deducting capital expenditures and changes in working capital. It is a key indicator of a company's financial health and value. The calculation formula and key parameters are as follows:

[0131] Core calculation formula

[0132] FCFF = Earnings before interest and taxes (EBIT) × (1-tax rate) + Depreciation and amortization - Capital expenditures - Changes in working capital.

[0133] The weighted average cost of capital (WACC) is the average cost of various types of capital, calculated by weighting them by their market value, reflecting the overall cost of financing. It is a core parameter in corporate valuations (such as the DCF model) and investment decisions (such as NPV analysis). Below is the WACC calculation formula, key parameter deriving methods, and examples:

[0134] Core calculation formula

[0135] WACC=(E / V×re)+(D / V×rd×(1-Tc))

[0136] E: Market value of equity capital (stock market value).

[0137] D: Market value of debt capital (current value of bonds or borrowings).

[0138] V=E+D: Market value of the enterprise’s total capital.

[0139] re: Cost of equity capital (the rate of return required by common shareholders).

[0140] rd: Cost of debt capital (pre-tax interest rate on bonds or borrowings).

[0141] Tc: corporate income tax rate.

[0142] Terminal value (TV) is a core concept in financial valuation, used to measure the ongoing value of an asset or business beyond a specific point in the future. It is based on the assumption that the business or asset will continue to operate and generate cash flow after the end of the forecast period and is a key component of the discounted cash flow model (DCF model).

[0143] In an embodiment of the present invention, determining a trading decision for an investment target based on the financial data, market performance score, quality factor QMJ indicator, and intrinsic value data of the investment target includes:

[0144] The financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target are input into the preset dynamic memory-focus synergy model to determine the research coefficient of the decision-making model, and the trading decision of the investment target is determined based on the subscription and redemption setting information and trading decision conditions.

[0145] In the embodiment of the present invention, the decision model is a buyer's extraordinary research model or a seller's extraordinary research model;

[0146] The financial data, market performance scores, quality factor QMJ indicators and intrinsic value data of the investment targets are normalized using a rolling window, with the window length W aligned with the financial reporting period;

[0147] Perform winsoring on extreme values ​​in the financial data, market performance score, quality factor QMJ indicator, and intrinsic value data of the investment target, using a preset quantile cutoff;

[0148] A dynamic memory gated temporal encoder with macroeconomic gating is used to perform nonlinear coupling analysis on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target;

[0149] The nonlinear coupling analysis results are processed in time channel and cross-sectional channel by a dual-channel adaptive feature focuser to generate the survey coefficients of the decision model, where:

[0150] Time channel: Analyze the contribution of the investment target's financial data, market performance score, quality factor QMJ indicator and intrinsic value data in the economic cycle;

[0151] Cross-sectional channel: Calculate the relative ranking of the investment target's financial data, market performance score, quality factor QMJ indicator and intrinsic value data within the industry.

[0152] In the process of obtaining the survey coefficients of the decision model, the embodiment of the present invention can realize automatic dynamic weight prediction and adjustment through the Dynamic Memory-Focus Mechanism (DMFCS):

[0153] Temporal feature extraction layer: Uses the DMGU unit (dynamic memory gated temporal encoder) with macroeconomic gating to perform nonlinear coupling analysis on financial data such as ROE stability and gross profit growth rate.

[0154] Dynamic weight generation layer: Generates time-varying weights (i.e., the survey coefficients of the decision model) through a dual-channel AFF (Adaptive Feature Focuser), where:

[0155] Time channel: Analyze the contribution of various financial data to the economic cycle (expansion / recession) defined by the NBER (National Bureau of Economic Research).

[0156] Cross-sectional channel: Calculate the Rank IC value of financial data within the industry and dynamically suppress the homogeneity factor.

[0157] Closed-loop feedback system: The posterior IC value of the screening result is used as a reinforcement learning reward signal, and the weight generator is continuously optimized through the PPO (Proximal Policy Optimization, deep reinforcement learning) algorithm.

[0158] For example, when the market is in a bear market, DMGU's macro-gating system detects a decline in GDP growth, triggering the AFF's defensive factor focus mode to increase the weight of the ROE stability factor by 30%. When the market shifts to a bull market, the growth rate of margin financing balances serves as an AFF cross-sectional channel enhancement signal, increasing the weight of the quality growth factor by 25%. As earnings reports approach, high-frequency weight updates (daily adjustments) are temporarily enabled. When a sudden change in earnings data of more than 10% is detected, the online update process can be initiated.

[0159] The dynamic memory-focusing collaborative system relies on the following parts:

[0160] ①Dynamic standardization module

[0161] - Use rolling window Z-score normalization, with the window length W aligned with the financial reporting cycle (W = 60 months)

[0162] -Winsorize extreme values ​​(99% quantile truncation) to prevent training divergence

[0163] -Introducing an adaptive scaling factor λ:

[0164] λt=1-0.5e -σt

[0165] Where σt is the volatility.

[0166] ②DMGU timing encoder

[0167] The Dynamic Memory Gated Temporal Encoder (DMGU) is an improved LSTM architecture designed specifically for financial time series data. Its core principle is to enhance traditional models through three innovative mechanisms: 1) a macro-gated forgetting mechanism, which integrates economic signals such as Treasury yields into a forgetting gate via a trainable matrix U_f, achieving economic cycle-aware memory decay; 2) a financial mutation-aware input gate, which automatically reduces the impact of abnormal data using an attenuation coefficient β (0.3-0.5) when detecting sharp fluctuations in key indicators such as ROE; and 3) an asymmetric state update, which uses differentiated update coefficients of 1.2 / 0.8 for positive / negative gross profit growth, respectively. By dynamically integrating macro-environmental signals with micro-financial characteristics, this architecture significantly improves its modeling capabilities for market transitions and financial report fluctuations.

[0168] ③AFF dual-channel focusing

[0169] Time channel: Based on the time series features encoded by LSTM (Long Short-Term Memory), the attention mechanism is used to calculate the historical contribution of each factor in different economic cycles (such as the expansion / recession phase defined by NBER) and generate time-varying weights. Time channel attention calculation:

[0170] α t T =softmax(q T tanh(W T h t ))

[0171] Among them, the parameter meaning and financial logic are shown in Table 1

[0172] Table 1

[0173]

[0174] Cross-sectional channel calculation: By standardizing industry groupings (such as calculating the relative ranking of factor values ​​within an industry), combined with the current market style (value / growth), the importance weights of factors within an industry are dynamically adjusted. The calculation formula is as follows:

[0175]

[0176] s(i) is the industry grouping, fi is the i-th financial data in the QMJ economic model, It is the average value of the financial data in the corresponding industry group.

[0177] In the embodiment of the present invention, the dynamic memory-focusing collaborative system needs to be trained and optimized. The process is as follows:

[0178] 1 Data preprocessing process

[0179] Data augmentation strategy:

[0180] Time series interpolation: cubic spline interpolation is used for the months between financial reports.

[0181] Adversarial Sample Generation: Synthesizing Market Crisis Data via WGAN-GP

[0182] 2 Joint Training Strategy

[0183] -Hybrid loss function design:

[0184] -IC loss: using the negative exponential form of Rank IC exp(-IC 21D )

[0185] - Market state classification: using FocalLoss with class balancing

[0186] -Smoothness constraint: limit the weight change rate of adjacent months ‖w t -w t-1 ‖ 2

[0187] 3 Online update mechanism

[0188] -The system also implements an online update mechanism. By setting the weight update calculation function, the retraining mechanism is triggered when the difference between the old and new weights is greater than the threshold, thereby improving the effectiveness and accuracy of training. The triggering condition is that the cosine similarity threshold is less than the preset value:

[0189] cos(w old ,w new )<0.3.

[0190] In an embodiment of the present invention, the stock trading decision is determined as follows: the stock is analyzed based on information such as the stock's basic information, financial indicator data, profit forecast (which can be data parsed from a profit forecast file uploaded by the user, or other information), company valuation (derived stock intrinsic value data), and BVIP summary (derived by integrating various indicator data). Factors such as the stock's win rate, expected rate of return, intrinsic value, and safety margin are considered to decide whether to hold a position, buy or sell the stock, and to determine the quantity and price of the purchase or sale. For example, if a stock has a high win rate, the expected rate of return reaches the set target, and the intrinsic value is higher than the current market price, a decision may be made to buy the stock, and the purchase quantity and price will be determined based on risk preference and funding conditions.

[0191] In an embodiment of the present invention, the determination of fund trading decisions is as follows: based on the fund subscription and redemption setting information (such as subscription method, subscription date, subscription amount, etc.) and the transaction decision condition setting information (such as valuation indicators, market conditions, etc.), combined with the fund's current holdings and investment objectives, the fund's trading decisions are generated. For example, when the valuation of a certain stock in a fund's investment portfolio reaches a preset selling condition (such as the market value of the individual stock is greater than a certain proportion of Qianpu's predicted net profit (FY1)), a triggering transaction decision is generated to determine the quantity and timing of selling the stock to adjust the fund's investment portfolio and achieve the fund's return target. At the same time, in terms of fund subscription and redemption, based on market conditions and the fund's funding needs, it is decided whether to subscribe or redeem the fund, as well as the amount and time of subscription or redemption.

[0192] In the embodiment of the present invention, the preset decision model is a buyer's extraordinary research model or a seller's extraordinary research model;

[0193] The buyer's extraordinary research model is expressed as:

[0194] LNMEETING i,m

[0195] =β0+β1SIZE i,m +β2TURN i,m +β3MOMEN i,m +β4ROA i,m

[0196] +∈ i,m (AIM)

[0197] Among them, LNMEETING i,m It represents the logarithm of the number of all investor surveys in the past 90 days for the i-th investment target at the end of the m-th month;

[0198] SIZE i,m : represents the logarithm of the circulating market value of the i-th investment target at the end of the m-th month;

[0199] TURN i,m : represents the average turnover rate of the i-th investment target at the end of the m-th month in the past 12 months;

[0200] MOMEN i,m : represents the cumulative stock return rate of the i-th investment target at the end of the m-th month, over a rolling 12-month period;

[0201] ROA i,m : represents the return on total assets (ROA) of the i-th investment target at the end of month m over a rolling 12-month period;

[0202] ∈ i,m (AIM): represents the value of the buy-side extraordinary survey of the i-th investment target at the end of month m;

[0203] β0, β1, β2, β3, and β4 are the buyer’s extraordinary research coefficients;

[0204] When ∈ i,m (AIM) is greater than the set buying threshold, or ∈ i,m When the ranking of (AIM) is higher than the preset ranking of all investment targets, confirm the purchase;

[0205] When ∈ i,m (AIM) is less than the set sell threshold, or ∈ i,m When the ranking of (AIM) is below the preset ranking of all investment targets, it is determined to sell;

[0206] The buyer's extraordinary research model is expressed as:

[0207] Log(1+TOT i,m

[0208] =α0+α1SIZE i,m +α2TURN i,m +α3MOMEN i,m

[0209] +∈ i,m (ATOT)

[0210] Among them, TOT i,m : represents the number of analysts covering the i-th investment target in the past 90 days at the end of the m-th month;

[0211] ∈ i,m (ATOT) represents the value of the sell-side extraordinary survey of the i-th investment target at the end of month m;

[0212] α0, α1, α2, and α3 are the seller’s extraordinary research coefficients;

[0213] When ∈ i,m When (ATOT) is greater than the set buying threshold, or ∈ i,m When the ranking of (AIM) is higher than the preset ranking of all investment targets, confirm the purchase;

[0214] When ∈ i,m When (ATOT) is less than the set selling threshold, or ∈ i,m When the ranking of (AIM) is below the preset ranking of all investment targets, it is determined to sell.

[0215] The decision model in the embodiments of the present invention can be implemented by programming a Python-based buy-side or sell-side unconventional research model, or other decision tree or machine learning algorithm. The algorithm inputs various analytical data about the stock (such as win rate, expected rate of return, intrinsic value, etc.) and outputs a trading decision (buy, sell, or hold). After the trading decision is determined, the buy or sell quantity and price are calculated, and the trading instruction is submitted.

[0216] Taking fund subscriptions and redemptions as an example, the decision-making process is explained: Pre-set information (such as subscription method, subscription date, and subscription amount) and trading decision conditions (such as valuation indicators and market conditions) are combined with the fund's current holdings and investment objectives to generate a fund trading decision. For example, when the valuation of a stock in a fund's portfolio reaches a preset sell condition (such as the individual stock's market capitalization exceeding a certain percentage of the projected net profit (FY1)), the system triggers a trading decision, determining the quantity and timing of the stock's sale to adjust the fund's portfolio and achieve the fund's return target. Simultaneously, regarding fund subscriptions and redemptions, the system determines whether to subscribe or redeem the fund, as well as the amount and timing of the subscription or redemption, based on market conditions and the fund's funding needs.

[0217] In an embodiment of the present invention, the value investment and trading decision-making method based on behavioral finance further includes: obtaining investment research related information; the investment research related information at least includes: company action event information;

[0218] When determining the trading decision for an investment target, a trading decision analysis is conducted based on the company's action event information, the strategic decision and operational dynamics of the investment target, and the investment target.

[0219] When there is a demand for stock investment, the transaction decision task is triggered by pre-setting transaction trigger conditions, and the corresponding investment target (such as selected stocks or funds) is determined for analysis. Preferably, in the embodiment of the present invention, investment research related information can be obtained for fund trading decisions. Investment research related information may include company action event information. In the embodiment of the present invention, company action event information is obtained by querying the company action event database table. Through SQL query statements, the corresponding company action event information is filtered and displayed according to the query conditions entered by the user (such as event category, event name, etc.).

[0220] like Figure 2 As shown, a value investment and trading decision-making system based on behavioral finance includes:

[0221] Investment research task management, which is used to obtain financial data of each investment target that triggers the trigger condition in response to the preset transaction trigger condition. The financial data includes: basic information, financial indicator data, volume and price indicator data, and shareholder transaction data;

[0222] An individual stock research module is used to determine the market performance score of an investment target based on the financial indicator data and the volume and price indicator data in the financial data of the investment target and the preset weights;

[0223] The individual stock research module is also used to determine the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target;

[0224] The individual stock research module is further configured to determine intrinsic value data of an investment target based on the financial data of the investment target using a discounted cash flow model, wherein the intrinsic value data indicates whether the investment value of the investment target is underestimated or overestimated;

[0225] A transaction management module is used to determine transaction decisions for investment targets based on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target. The transaction decisions may include one or more of the following: whether to buy, whether to sell, whether to hold a position, the timing of buying or selling, and the quantity and price of buying or selling.

[0226] In an embodiment of the present invention, the transaction management module is used to input the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target into a preset decision model, and determine the transaction decision for the investment target based on the subscription and redemption setting information and transaction decision conditions.

[0227] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0228] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A value investment and trading decision-making method based on behavioral finance, characterized by: include: In response to a preset transaction trigger condition, obtaining financial data of each investment target that initiated the trigger condition, the financial data including: basic information, financial indicator data, volume and price indicator data, and shareholder transaction data; Determine the market performance score of the investment target based on the financial indicator data and the quantity and price indicator data in the financial data of the investment target and the preset weights; Determine the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target; Determining intrinsic value data of the investment target based on the financial data of the investment target using a discounted cash flow model, wherein the intrinsic value data indicates whether the investment value of the investment target is underestimated or overestimated; The trading decision of the investment target is determined based on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target. The trading decision may include one or more of the following: whether to buy, whether to sell, whether to hold a position, the timing of buying or selling, and the quantity and price of buying or selling.

2. The value investment and trading decision-making method based on behavioral finance according to claim 1, characterized in that: The transaction triggering condition includes at least one of the following: The scheduled time or period arrives; The change in the quality of the individual stocks of the investment target meets the first condition; The change in the valuation of the investment target satisfies the second condition; The change in the behavioral signal of the investment target satisfies the third condition; The expected return change of the investment target satisfies the fourth condition; The basic information includes at least one of the following: Code, name, sector, listing date, and industry classification; The financial indicator data includes at least one of the following: Profit before interest and taxes, operating income, net profit, return on equity (ROE), return on total assets (ROA), gross profit margin; The quantity and price indicator data includes at least one of the following: Trading volume, closing price, opening price, highest price, lowest price; The shareholder transaction data includes: the cumulative increase or decrease in shareholdings by shareholders and executives within a predetermined period of time.

3. The value investment and trading decision-making method based on behavioral finance according to claim 2, characterized in that: Determine the market performance score of the investment target based on the financial indicator data and quantity and price indicator data in the financial data of the investment target, as well as the preset weights, including: According to the rise and fall of each indicator data in the financial indicator data and the quantity and price indicator data, numerical statistics are performed, wherein the rise is recorded as the first value and the fall is recorded as the second value; The market performance score of the investment target is calculated by taking the weighted sum of the numerical value of each indicator data and the weight of each indicator number.

4. The value investment and trading decision-making method based on behavioral finance according to claim 2, characterized in that: Determine the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target, including: Calculate the rolling 12-month gross profit per unit of total assets by dividing the gross profit and total assets data in the financial indicator data; The growth rate of ROE for rolling 12 months is calculated by comparing ROE of different time periods; The volatility of the ROE for the rolling 12 months is calculated by comparing the ROE of different time periods.

5. The value investment and trading decision-making method based on behavioral finance according to claim 2, characterized in that: Using a discounted cash flow model, based on the financial data of the investment target, determine the intrinsic value data of the investment target, wherein the intrinsic value data reflects whether the investment value of the investment target is underestimated or overestimated, including: By inputting the predicted future free cash flow FCFF, the predicted future weighted average cost of capital WACC, the year-on-year growth rate of operating income, the profit margin before interest and taxes, and the operating income / invested capital less cash and equivalents, and using the cash flow discount model, the intrinsic value data Z of the investment target is calculated according to the following formula: Z=∑[FCFF(t) / (1+WACC)^t]+final value / (1+WACC)^n; Where t represents the year of the forecast period and n represents the last year of the forecast period.

6. The value investment and trading decision-making method based on behavioral finance according to any one of claims 2 to 5, characterized in that: Determine trading decisions for investment targets based on the target's financial data, market performance score, quality factor (QMJ) indicator, and intrinsic value data, including: Input the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target into the preset dynamic memory-focus synergy model to determine the research coefficient of the decision-making model; Determine trading decisions based on investment objectives based on subscription and redemption setting information and trading decision conditions.

7. The value investment and trading decision-making method based on behavioral finance according to claim 6, characterized in that: The decision model is a buyer's extraordinary research model or a seller's extraordinary research model; The financial data, market performance scores, quality factor QMJ indicators and intrinsic value data of the investment targets are normalized using a rolling window, with the window length W aligned with the financial reporting period; Perform winsoring on extreme values ​​in the financial data, market performance score, quality factor QMJ indicator, and intrinsic value data of the investment target, using a preset quantile cutoff; A dynamic memory gated temporal encoder with macroeconomic gating is used to perform nonlinear coupling analysis on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target; The nonlinear coupling analysis results are processed in time channel and cross-sectional channel by a dual-channel adaptive feature focuser to generate the survey coefficients of the decision model, where: Time channel: Analyze the contribution of the investment target's financial data, market performance score, quality factor QMJ indicator and intrinsic value data in the economic cycle; Cross-sectional channel: Calculate the relative ranking of the investment target's financial data, market performance score, quality factor QMJ indicator and intrinsic value data within the industry.

8. The value investment and trading decision-making method based on behavioral finance according to claim 7, characterized in that: The buyer's extraordinary research model is expressed as: LNMEETING i,m =β0+β1SIZE i,m +β2TURN i,m +β3MOMEN i,m +β4ROA i,m +∈ i,m (AIM) Among them, LNMEETING i,m It represents the logarithm of the number of all investor surveys in the past 90 days for the i-th investment target at the end of the m-th month; SIZE i,m : represents the logarithm of the circulating market value of the i-th investment target at the end of the m-th month; TURN i,m : represents the average turnover rate of the i-th investment target at the end of the m-th month in the past 12 months; MOMEN i,m : represents the cumulative stock return rate of the i-th investment target at the end of the m-th month, over a rolling 12-month period; ROA i,m : represents the return on total assets (ROA) of the i-th investment target at the end of month m over a rolling 12-month period; ∈ i,m (AIM): represents the value of the buy-side extraordinary survey of the i-th investment target at the end of month m; β0, β1, β2, β3, and β4 are the buyer’s extraordinary research coefficients; When ∈ i,m (AIM) is greater than the set buying threshold, or ∈ i,m When the ranking of (AIM) is higher than the preset ranking of all investment targets, confirm the purchase; When ∈ i,m (AIM) is less than the set sell threshold, or ∈ i,m When the ranking of (AIM) is below the preset ranking of all investment targets, it is determined to sell; The buyer's extraordinary research model is expressed as: Log(1+TOT i,m )=α0+α1SIZE i,m +α2TURN i,m +α3MOMENT i,m +∈ i,m (ASSET) Among them, TOT i,m : represents the number of analysts covering the i-th investment target in the past 90 days at the end of the m-th month; ∈ i,m (ATOT) represents the value of the sell-side extraordinary survey of the i-th investment target at the end of month m; α0, α1, α2, and α3 are the seller’s extraordinary research coefficients; When ∈ i,m When (ATOT) is greater than the set buying threshold, or ∈ i,m When the ranking of (AIM) is higher than the preset ranking of all investment targets, confirm the purchase; When ∈ i,m When (ATOT) is less than the set selling threshold, or ∈ i,m When the ranking of (AIM) is below the preset ranking of all investment targets, it is determined to sell.

9. The value investment and trading decision-making method based on behavioral finance according to claim 1, characterized in that: Also includes: Obtain investment research related information; The investment research related information includes at least: company action event information; When determining the trading decision for an investment target, a trading decision analysis is conducted based on the company's action event information, the strategic decision and operational dynamics of the investment target, and the investment target.

10. A value investment and trading decision-making system based on behavioral finance, characterized by: include: Investment research task management, which is used to obtain financial data of each investment target that triggers the trigger condition in response to the preset transaction trigger condition. The financial data includes: basic information, financial indicator data, volume and price indicator data, and shareholder transaction data; An individual stock research module is used to determine the market performance score of an investment target based on the financial indicator data and the volume and price indicator data in the financial data of the investment target and the preset weights; The individual stock research module is also used to determine the quality factor QMJ indicator based on the financial indicator data in the financial data of the investment target; The individual stock research module is further configured to determine intrinsic value data of an investment target based on the financial data of the investment target using a discounted cash flow model, wherein the intrinsic value data indicates whether the investment value of the investment target is underestimated or overestimated; A transaction management module is used to determine transaction decisions for investment targets based on the financial data, market performance score, quality factor QMJ indicator and intrinsic value data of the investment target. The transaction decisions may include one or more of the following: whether to buy, whether to sell, whether to hold a position, the timing of buying or selling, and the quantity and price of buying or selling.

Images