Risk prediction method and device, equipment and storage medium

Abstract

The invention provides a risk prediction method and device, equipment and a storage medium, and the method comprises the steps: obtaining a fund identifier and fund features of a target fund, matching the fund features with a pre-constructed risk feature database according to the fund identifier, and obtaining a direct risk value corresponding to the target fund; obtaining a corresponding security identifier and security features according to the position information of the target fund, matching the security features with a pre-constructed risk feature database according to the security identifier, and obtaining an indirect risk value corresponding to the position information of the target fund; determining a comprehensive risk value corresponding to the target fund according to the direct risk value and the indirect risk value, and determining whether to give an early warning to the target fund according to the comprehensive risk value; the fund risk identification method and device are used for improving comprehensiveness and accuracy of fund risk identification through cross-dimensional linkage monitoring of fund and security risks.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a risk prediction method, apparatus, device, and storage medium. Background Technology

[0002] Public fund risk prediction is based on the forecast and judgment of risks such as market, credit and liquidity. As the fund size continues to expand, investment behavior is affected by multiple factors such as financial market fluctuations, changes in corporate fundamentals and public opinion, making the need for risk management increasingly urgent. In current wealth management services, the introduction, evaluation, recommendation and tracking of public funds have been generally established, but risk identification still mainly relies on traditional methods, which makes it difficult to respond to market dynamics and changes in public opinion in a timely manner, resulting in investors and regulators facing a situation of delayed risk discovery and insufficient response.

[0003] Currently, the risk monitoring technologies commonly used in the industry for public funds mainly include the following methods: First, based on the analysis of financial data and net asset value indicators, abnormal performance fluctuations are identified by setting thresholds and comparing historical data with similar products; second, compliance and trading rules are relied upon, combined with fund announcements, periodic reports, and regulatory information, and compliance warnings are achieved through keyword searches. However, existing risk warnings are generally delayed, relying heavily on periodic data or post-event announcements, making it difficult to cope with sudden market events and causing investors to miss the best opportunity to deal with risks. Summary of the Invention

[0004] This application provides a risk prediction method, apparatus, device, and storage medium, which improves the comprehensiveness and accuracy of fund risk identification through cross-dimensional linkage monitoring of fund and securities risks, enables early warning of fund risks, and provides strong data support for investment decisions and risk management.

[0005] Firstly, this application provides a risk prediction method, including:

[0006] Obtain the fund identifier and fund characteristics of the target fund; match the fund characteristics with a pre-built risk characteristic database based on the fund identifier to obtain the direct risk value corresponding to the target fund.

[0007] Based on the target fund's holdings information, obtain the corresponding security identifier and security characteristics, and match the security characteristics with a pre-built risk characteristic database to obtain the indirect risk value corresponding to the target fund's holdings information;

[0008] Based on the direct risk value and the indirect risk value, determine the comprehensive risk value corresponding to the target fund, and determine whether to issue an early warning for the target fund based on the comprehensive risk value.

[0009] In one or more possible embodiments, the risk characteristic database is constructed as follows:

[0010] Collect raw information on funds and securities, including fund names, fund net asset values, fund financial reports, securities trading information, financial statements of the companies corresponding to the securities, news information, social media sentiment, and macroeconomic market data;

[0011] From the original information, obtain the risk characteristics corresponding to each fund and security; the risk characteristics include at least one of the following: fund type, net asset value volatility, Sharpe ratio, maximum drawdown, securities trading volume, securities trading price, financial indicators, degree of negative public opinion, and valuation deviation;

[0012] Based on a preset risk value determination method, the risk value corresponding to each risk characteristic is determined.

[0013] The risk characteristic database is formed by associating the identifiers of funds and securities with their corresponding risk characteristics and risk values.

[0014] After the risk feature database is constructed, the following steps are also included:

[0015] When the conditions for updating the risk characteristic database are met, the risk characteristic database is updated. In one or more possible embodiments, the degree of negative public opinion is obtained in the following way:

[0016] Natural language processing technology is used to perform semantic analysis on collected news information and social media sentiment to obtain structured data on each fund and security.

[0017] The structured data is analyzed using sentiment analysis algorithms to determine the degree of negative public opinion for each fund and security.

[0018] In one or more possible embodiments, determining whether to issue an early warning to the target fund based on the comprehensive risk value includes:

[0019] The overall risk value is compared with multiple preset risk level thresholds;

[0020] Based on the comparison results, the risk level of the target fund is determined;

[0021] When the risk level reaches or exceeds the preset warning level, a warning message containing the risk level is generated and output.

[0022] In one or more possible embodiments, the step of obtaining the corresponding security identifier and security characteristics based on the target fund's holdings information, and matching the security characteristics with a pre-built risk characteristic database based on the security identifier to obtain the indirect risk value corresponding to the target fund's holdings information, includes:

[0023] Based on the target fund's holdings information, obtain all securities held by the target fund and the holding ratio of each security;

[0024] The security characteristics of each security are matched with the risk characteristic database to obtain the individual risk value of each security;

[0025] Based on the holding ratio of each security, the individual risk values ​​are weighted and aggregated to obtain the indirect risk value.

[0026] In one or more possible embodiments, the method further includes:

[0027] Once the trading behavior of the target fund is obtained and it is determined that the trading behavior exceeds the preset trading rules, an early warning is issued to the target fund.

[0028] Secondly, this application also provides a risk prediction device, comprising:

[0029] The direct risk value determination module is used to obtain the fund identifier and fund characteristics of the target fund, and match the fund characteristics with a pre-built risk characteristic database based on the fund identifier to obtain the direct risk value corresponding to the target fund;

[0030] The indirect risk value determination module is used to obtain the corresponding security identifier and security characteristics based on the target fund's holding information, and match the security characteristics with a pre-built risk characteristic database based on the security identifier to obtain the indirect risk value corresponding to the target fund's holding information.

[0031] The early warning module determines the comprehensive risk value corresponding to the target fund based on the direct risk value and the indirect risk value, and determines whether to issue an early warning to the target fund based on the comprehensive risk value.

[0032] Thirdly, this application also provides an electronic device, the electronic device comprising:

[0033] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the methods in the first aspect.

[0034] Fourthly, this application also provides a computer storage medium storing a computer program for causing a computer to perform any of the methods described in the first aspect.

[0035] According to the risk prediction method, device, equipment and storage medium provided in this application, the comprehensiveness and accuracy of fund risk identification can be improved through cross-dimensional linkage monitoring of fund and securities risks, early warning of fund risks can be achieved, and strong data support can be provided for investment decision-making and risk management. Attached Figure Description

[0036] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application, and do not constitute an undue limitation of this application.

[0037] Figure 1 This is a schematic diagram illustrating an application scenario of a risk prediction method provided according to an embodiment;

[0038] Figure 2 This is a flowchart of a risk prediction method provided according to an embodiment;

[0039] Figure 3 This is a block diagram of a risk prediction system according to an embodiment;

[0040] Figure 4 This is a schematic diagram of a risk prediction device according to an embodiment;

[0041] Figure 5 This is a schematic diagram of an electronic device according to an embodiment;

[0042] Figure 6 This is a schematic diagram of a computer-readable storage medium provided according to an embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0044] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0045] Furthermore, in the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0046] For ease of understanding, the risk prediction method provided by the embodiments of this application will be described in detail below with reference to the accompanying drawings:

[0047] like Figure 1 The diagram shown illustrates an application scenario of a risk prediction method provided in this embodiment. The diagram includes: network 10, server 20, and storage 30.

[0048] The description in this application focuses on a single server only. However, those skilled in the art should understand that the illustrated network 10, server 20, and memory 30 are intended to illustrate the operation of the electronic devices, servers, and memory involved in the technical solutions of this application. The detailed description of a single server and memory is at least for ease of explanation and does not imply any limitation on the number, type, or location of servers. It should be noted that adding additional modules to or removing individual modules from the illustrated environment will not change the underlying concept of the exemplary embodiments of this application. Furthermore, although detailed descriptions are provided for ease of explanation... Figure 1 The diagram shows a bidirectional arrow from memory 30 to server 20, but those skilled in the art will understand that the sending and receiving of the aforementioned data also needs to be achieved through network 10.

[0049] It should be noted that the memory in the embodiments of this application can be, for example, a cache system, hard disk storage, memory storage, etc. Furthermore, the risk prediction method proposed in this application is not only applicable to… Figure 1 The application scenarios shown can also be used in other possible application scenarios, and the embodiments of this application do not impose any limitations.

[0050] Current risk predictions for mutual funds are based on forecasts and judgments of risks such as market, credit, and liquidity. As the size of funds continues to expand, investment behavior is affected by multiple factors such as financial market fluctuations, changes in corporate fundamentals, and public opinion, making the need for risk management increasingly urgent. In current wealth management services, the introduction, evaluation, recommendation, and tracking of mutual funds have been widely established, but risk identification still mainly relies on traditional methods, making it difficult to respond promptly to market dynamics and changes in public opinion. This results in investors and regulators facing situations of delayed risk discovery and inadequate response.

[0051] Currently, the risk monitoring technologies commonly used in the industry for public funds mainly include the following methods: First, based on the analysis of financial data and net asset value indicators, abnormal performance fluctuations are identified by setting thresholds and comparing historical data with similar products; second, compliance and trading rules are relied upon, combined with fund announcements, periodic reports, and regulatory information, and compliance warnings are achieved through keyword searches. However, existing risk warnings are generally delayed, relying heavily on periodic data or post-event announcements, making it difficult to cope with sudden market events and causing investors to miss the best opportunity to deal with risks.

[0052] Based on the above problems, this application provides a risk prediction method, such as... Figure 2 As shown, it includes:

[0053] Step 201: Obtain the fund identifier and fund characteristics of the target fund; match the fund characteristics with a pre-built risk characteristic database based on the fund identifier to obtain the direct risk value corresponding to the target fund.

[0054] In one or more possible embodiments, basic information about the target fund can be obtained from an internal database. This basic information includes the fund's identifier (a unique identifier for the fund), such as the fund code or full name, as well as fund characteristics reflecting the fund's operational status. Fund characteristics include, but are not limited to, performance indicators such as net asset value volatility, Sharpe ratio, and maximum drawdown, and compliance indicators such as portfolio concentration and trading turnover. These fund characteristics can be obtained in real-time from official channels such as exchanges and clearinghouses. Then, using the target fund identifier as an index key, the obtained fund characteristics are matched against a pre-built risk characteristic database. The risk feature database stores risk features and their corresponding risk value calculation rules. For example, for the feature "net asset value volatility," the database contains the following calculation rules: the risk value is 20 when the net asset value volatility is below 3%, 50 between 3% and 8%, and 80 above 8%. Then, based on the matched features and corresponding calculation rules of the target fund, the database automatically calculates the corresponding risk score for each fund feature. Finally, through weighted averaging or other aggregation algorithms, these risk scores belonging to different dimensions are combined into a direct risk value. This direct risk value centrally reflects the risk level caused by the fund's own factors.

[0055] Step 202: Obtain the corresponding security identifier and security characteristics based on the target fund's holding information; match the security characteristics with a pre-built risk characteristic database based on the security identifier to obtain the indirect risk value corresponding to the target fund's holding information.

[0056] In one or more possible embodiments, based on the target fund's holdings information, all securities held by the target fund and the holding ratio of each security are obtained; the security characteristics of each security are matched with the risk characteristic database to obtain the individual risk value of each security; based on the holding ratio of each security, the individual risk values ​​are weighted and aggregated to obtain the indirect risk value; specifically, firstly, the latest holdings report of the target fund is obtained, and then the identification information of all securities included in the investment portfolio and their corresponding holding ratios are obtained based on the latest holdings report of the target fund; the above holdings data usually comes from the fund's periodic reports (such as quarterly reports, semi-annual reports, and annual reports), and can also be obtained in real time; the above securities The identifiers include information that uniquely identifies a security, such as stock codes and bond ISIN codes. After obtaining a complete list of securities (identification information of all securities included in the target fund and their corresponding holding ratios), a batch query is initiated in the risk characteristic database using the security identifiers. The risk characteristic database then returns a quantified individual risk value generated for each security. After obtaining the individual risk values ​​of all securities in the portfolio, a weighted calculation is performed according to the holding ratio of each security in the target fund. Specifically, the individual risk value of each security is multiplied by its weight in the total fund investment, and then all weighted risk values ​​are summed to obtain the indirect risk value reflecting the overall holding risk of the fund. The calculation formula is as follows:

[0057] Indirect risk value = Σ(holding ratio of security i × individual risk value of security i);

[0058] For example, if a target fund holds three securities, A, B, and C, with holding ratios of 40%, 35%, and 25% respectively, and the risk values ​​for each security obtained from the risk characteristic database are 65, 80, and 50 respectively, then the indirect risk value of the target fund is calculated as: 40% × 65 + 35% × 80 + 25% × 50 = 26 + 28 + 12.5 = 66.5. This indirect risk value accurately reflects the associated risk level borne by the target fund due to its securities holdings. Through this risk transmission mechanism based on holding weights, a precise mapping from individual security risk to the overall fund risk is achieved. This allows the fund's risk assessment to not only consider its own operational status but also fully incorporate the quality factors of the underlying assets, thereby constructing a more comprehensive risk monitoring method and improving the accuracy of risk prediction for the target fund.

[0059] Step 203: Determine the comprehensive risk value corresponding to the target fund based on the direct risk value and the indirect risk value, and determine whether to issue a warning to the target fund based on the comprehensive risk value.

[0060] In one or more possible embodiments, the direct risk value and the indirect risk value are first combined into a comprehensive risk value using a weighted fusion algorithm. This comprehensive risk value reflects a holistic consideration of the target fund's own operational risks and portfolio-related risks. Specifically, differentiated weighting schemes can be set for different types of funds. For example, equity funds may be given a higher weight for indirect risk values ​​(e.g., β = 0.6), while for money market funds, considering that their investment targets are mainly short-term monetary instruments, more attention will be paid to direct risks (e.g., liquidity risk, compliance risk, etc.). Therefore, a weighting ratio of α = 0.7 may be set for the direct risk value. Different weights are used for the direct and indirect risk values ​​corresponding to different types of target funds to ensure that the final risk prediction results are highly consistent with the characteristics of different types of fund products.

[0061] In one or more possible embodiments, after calculating the comprehensive risk value, different comprehensive risk values ​​will correspond to different risk levels. Multiple risk level thresholds, ranging from level one to level three, can be preset. These thresholds are not fixed but dynamically adjusted based on multiple dimensions such as fund type, overall market volatility, regulatory requirements, and user risk preferences. During periods of severe market volatility, the thresholds for each level can be lowered to enhance the sensitivity of risk monitoring and ensure adaptability to complex and ever-changing market environments. Then, the calculated comprehensive risk value is matched in real-time with these dynamic thresholds to accurately determine the fund's current risk level. The higher the level, the greater the fund's risk. For example, level one indicates a low risk level, meaning the fund is operating normally and only requires routine monitoring; the risk level threshold for level one is 30. Level two indicates... A medium risk level indicates that the fund is showing signs of risk and requires closer monitoring. The risk threshold for level two is 60. Level three indicates a high risk level, meaning the fund faces severe risk and an immediate warning must be triggered. The risk threshold for level three is 80. For example, if the calculated comprehensive risk value of the target fund is 70, it exceeds the risk threshold for level two. If the preset warning level or risk level of this application is level two, a warning message containing the level two risk level will be output. In addition, more risk levels can be set, which will not be listed here. Alternatively, a simpler method can be set to determine whether to issue a warning to the target fund based on the comprehensive risk value. For example, the threshold for issuing an alarm can be preset to 80. When the calculated comprehensive risk value of the target fund is greater than or equal to 80, an alarm will be issued directly.

[0062] In one or more possible embodiments, when a fund's risk level is identified as reaching or exceeding a preset warning level, an alarm will be issued immediately, and a warning report containing core information such as the target fund's basic information, risk level, and main sources of risk will be generated. Targeted handling suggestions can also be provided. For example, for a high-risk warning caused by the financial deterioration of heavily invested securities, operational guidelines such as "suggest checking the holding ratio" and "paying attention to subsequent announcements of relevant securities" will be given. The generated warning information can also be distributed through various channels, including visually highlighting it on a monitoring screen, sending push notifications to the mobile terminals of relevant responsible personnel, and generating pending tasks in the corresponding system.

[0063] In one or more possible embodiments, the risk prediction method further includes: acquiring the trading behavior of the target fund, and issuing an early warning to the target fund after determining that the trading behavior exceeds preset trading rules; specifically, a continuous data acquisition mechanism can be established to acquire the target fund's transaction flow data and holding change information in real time by connecting to data sources such as securities exchanges and fund custodians; these data contain complete transaction records, including key fields such as transaction time, security code, buy / sell direction, transaction price, and transaction quantity, as well as daily updated holding details and ratio data; after obtaining real-time transaction data, it is determined whether the trading behavior exceeds the preset trading rules, and if so, an early warning is issued to the target fund; the preset trading rules include multiple dimensions, for example, the user sets that the holding ratio of a single security of the target fund shall not exceed 10%, the same fund shall not buy or sell the same security more than 3 times in a single trading day, and the industry concentration shall not exceed 30% of the fund's net asset value, etc. The above trading rules can be flexibly configured according to regulatory requirements and the fund manager's risk policy; for example, when it is found that the target fund's holding ratio of a single stock reaches 12%, it is identified that the trading behavior violates the preset trading rules.

[0064] In one or more possible embodiments, the aforementioned preset trading rules not only support checking whether a single transaction is compliant, but also analyze trading sequences using statistical methods to identify potential abnormal patterns and determine whether the preset trading rules are exceeded. For example, if it is detected that a fund engages in buy-then-sell reversal trading on the same stock for five consecutive trading days, and the cumulative trading volume exceeds 20% of the stock's average daily trading volume, then it is determined that the aforementioned preset trading rules have been violated and an abnormal trading pattern is identified. When it is determined that the trading behavior exceeds the preset trading rules, a tiered early warning mechanism can also be activated, dividing the warning into multiple levels based on the nature and severity of the illegal trading behavior. For example, for minor violations, such as a trading frequency slightly exceeding the threshold, a warning is generated and logged; for major violations, such as a holding ratio seriously exceeding the limit, a high-level warning is immediately triggered. The warning information includes a specific description of the violation, such as "Fund A's holding ratio of stock B has reached 12%, exceeding the 10% limit." Through this method, which includes data acquisition, rule comparison, anomaly identification, and tiered warnings, violations in fund trading can be effectively identified, ensuring that fund operations comply with regulations and preventing trading risks. It can detect both explicit rule violations and potential abnormal trading patterns, greatly improving the comprehensiveness and accuracy of trading behavior monitoring.

[0065] In one or more possible embodiments, the risk characteristic database is constructed as follows: Raw information of funds and securities is collected, including fund name, fund net asset value, fund financial reports, securities trading information, financial statements of the corresponding companies, news information, social media sentiment, and macroeconomic market data; risk characteristics are obtained from the raw information for each fund and security; the risk characteristics include at least one of the following: fund type, net asset value volatility, Sharpe ratio, maximum drawdown, securities trading volume, securities trading price, financial indicators, degree of negative public opinion, and valuation deviation; a risk value is calculated for each risk characteristic based on a preset risk value calculation method; and the fund and security identifiers, corresponding risk characteristics, and corresponding... Risk values ​​are correlated to form a risk characteristic database. Specifically, raw information is first obtained from different data providers and public channels. Fund-related data is mainly obtained from official disclosures by fund companies, public information from exchanges, and third-party financial service institutions, including basic fund information, daily net asset value data, quarterly and annual financial reports, etc. Securities-related data comes from real-time trading records of stock exchanges, financial report announcements of listed companies, and public information from authoritative financial media. Social media sentiment is obtained by using web crawling and natural language processing technologies to capture and analyze public information such as news, forums, and social media on the Internet in real time. Macro market data includes authoritative information such as economic indicators released by the National Bureau of Statistics and monetary policy reports from the central bank.

[0066] After collecting the raw information, a keyword filtering method is also included. Specifically, the acquired raw information is initially screened based on a preset keyword list. Only when the raw information contains at least one keyword is further feature extraction performed to improve the accuracy of capture. For example, the preset keyword list may include core terms related to fund and securities risks, such as "financial fraud," "regulatory investigation," "major losses," "debt default," "performance reversal," "liquidity crisis," "rating downgrade," and "litigation disputes." When a new piece of public opinion information is collected, such as a news report titled "A listed company is under investigation for financial fraud," which contains the keyword "financial fraud," the information is deemed valid and enters the subsequent feature extraction stage. Conversely, if the information content is "A company's new product launch was a complete success," and does not contain any preset keywords, it is directly filtered and does not enter the subsequent processing flow. Through this keyword-based preliminary screening mechanism, the interference of irrelevant information can be significantly reduced, improving the accuracy and efficiency of risk information processing.

[0067] In one or more possible embodiments, after keyword filtering of the original information, the feature extraction stage is initiated. The extracted risk features include at least one of the following: fund type, net asset value volatility, Sharpe ratio, maximum drawdown, securities trading volume, securities trading price, financial indicators, degree of negative sentiment, and valuation deviation. For example, for a fund, which is an equity fund, volatility is calculated based on the fund's net asset value data over the past 20 trading days, and the Sharpe ratio is calculated by dividing the difference between the fund's annual return and the risk-free rate by the standard deviation of the return. For a security, the corresponding risk features are mainly extracted from the trading records, such as the number of transactions within a business day and the price of each transaction. The degree of negative sentiment, including the negative sentiment of the fund and the negative sentiment of the security, is analyzed using natural language processing technology to identify the degree of negative sentiment. The valuation deviation feature is calculated by comparing the difference between the current price-to-earnings ratio and the industry average price-to-earnings ratio.

[0068] In one or more possible embodiments, a different risk value determination method is preset for each risk characteristic, and the specific process is as follows:

[0069] Risk values ​​corresponding to fund types: The risk value for the first fund of type A is set at 10, and the risk value for the second fund of type B is set at 20.

[0070] The risk value corresponding to net asset value volatility is as follows: when the volatility is below 3%, the risk value is set to 20, indicating low risk; when the volatility is between 3% and 8%, the risk value is set to 50, indicating medium risk; when the volatility is above 8%, the risk value is set to 80, indicating high risk. For example, if the net asset value volatility of Fund 1 reaches 10%, the volatility risk value corresponding to Fund 1 is set to 80.

[0071] The risk values ​​corresponding to the Sharpe ratio are as follows: When the Sharpe ratio is greater than 1.5, the risk value is set at 20, indicating good risk-adjusted returns; when the Sharpe ratio is between 0.5 and 1.5, the risk value is set at 50; when the Sharpe ratio is less than 0.5, the risk value is set at 80. For example, if the second fund has a Sharpe ratio of 0.3, then the corresponding Sharpe ratio risk value for the second fund is 80.

[0072] The risk value corresponding to the maximum drawdown is set as follows: 20 for a maximum drawdown below 5%; 50 for a drawdown between 5% and 15%; and 80 for a drawdown exceeding 15%. For example, if a fund's maximum drawdown reaches 18%, its risk value is set to 80.

[0073] Risk Value Corresponding to Securities Trading Volume: Abnormal changes in trading volume (increased or decreased volume) are often a precursor to significant price changes or emerging risks. Therefore, the volume ratio indicator can be used to identify abnormal situations and determine the corresponding risk value. Volume Ratio = (Total Trading Volume on the Current Day / Average Total Trading Volume at the Same Time in the Past 5 Trading Days). This application uses the above formula to calculate the volume ratio. When the volume ratio is between 0.8 and 1.2, the risk value is set at 20; when the volume ratio is between 1.2 and 2 times or between 0.5 and 0.8 times, the risk value is set at 50; when the volume ratio exceeds 2 times or is less than 0.5 times, the risk value is set at 80. For example, if the volume ratio of a certain security reaches 3.5, its trading volume risk value is set at 80.

[0074] The risk value corresponding to the securities trading price is determined based on the price fluctuation range. If the cumulative increase or decrease is within ±5% over three consecutive trading days, the risk value is set to 20; if the cumulative increase or decrease is between ±5% and ±15%, the risk value is set to 50; and if the cumulative increase or decrease exceeds ±15%, the risk value is set to 80. For example, if a security has fallen by 20% cumulatively over three consecutive days, its price risk value is set to 80.

[0075] Risk values ​​corresponding to financial indicators: Taking the debt-to-equity ratio as an example for risk assessment. When the debt-to-equity ratio is below 50%, the risk value is set at 20; when it is between 50% and 70%, the risk value is set at 50; and when it exceeds 70%, the risk value is set at 80. For example, if a listed company's debt-to-equity ratio reaches 75%, then its financial indicator risk value is set at 80.

[0076] The risk value corresponding to the degree of negative public opinion is as follows: when the sentiment score is higher than -0.2, the risk value is set to 20; when the score is between -0.2 and -0.6, the risk value is set to 50; and when the score is lower than -0.6, the risk value is set to 80. For example, if the sentiment score of a certain securities' public opinion information is -0.8, then its negative public opinion risk value is set to 80.

[0077] The risk values ​​corresponding to valuation deviations are as follows: A risk value of 20 is set when the deviation is within ±10%; a risk value of 50 is set when the deviation is between ±10% and ±30%; and a risk value of 80 is set when the deviation exceeds ±30%. For example, if the valuation deviation of a security reaches 40%, its valuation deviation risk value is set to 80. For instance, this application can use a combination of the price-to-earnings ratio method and the historical quantile method to calculate the degree of valuation deviation of a stock. First, the method for determining the fair value (i.e., intrinsic value) is as follows:

[0078] Method 1 (Industry Comparison): Company A's earnings per share are 1.5 yuan, and the industry average price-to-earnings ratio is 20; fair value 1 = 1.5 yuan × 20 = 30 yuan.

[0079] Method 2 (Historical Analysis): Company A's current P / E ratio is 35, and the historical median P / E ratio of Company A over the past 5 years is 25; fair value 2 = 1.5 yuan × 25 = 37.5 yuan.

[0080] Comprehensive reasonable value: Take the average of the results of Method 1 and Method 2, or assign different weights, to calculate the final reasonable value; assume that the reasonable value obtained by taking the average is (30+37.5) / 2 = 33.75 yuan.

[0081] Then obtain the current market price, assuming Company A's current stock price is 45 yuan.

[0082] Finally, the following formula is used to calculate the degree of valuation deviation: Degree of valuation deviation = (45-33.75) / 33.75×100%≈33.3%; Based on the calculated degree of valuation deviation of the stock, which is 33.3% exceeding ±30%, the corresponding risk value is determined to be 80; that is, the risk value corresponding to the valuation deviation risk characteristic of this stock is 80 points.

[0083] It is worth noting that the calculation method for the risk characteristics and the corresponding range of risk values ​​given in this application are not fixed and can be adjusted according to user habits, market environment, regulatory requirements or user risk preferences, etc., which will not be elaborated here.

[0084] Finally, a relationship is established using fund code (fund identifier) ​​and security code (fund identifier) ​​as the primary key (index) for storage in the risk feature database. Each fund or security corresponds to multiple risk features, including fields such as feature type, feature value, risk score, and update time. Through an efficient indexing mechanism, fast queries based on fund or security identifiers are supported. At the same time, to ensure data quality, data validation rules can be set to mark outliers and conduct manual review, ensuring the accuracy and reliability of the data entering the database.

[0085] In one or more possible embodiments, after determining that the risk feature database has been constructed, the method further includes: updating the risk feature database when the conditions for updating the risk feature database are met; specifically, it includes the following process:

[0086] Update triggers primarily include two methods: timed triggers and event triggers. Timed triggers are set to perform a full update on the first trading day of each month, re-collecting complete data from the previous month. Event triggers, on the other hand, initiate updates immediately when specific circumstances occur, including situations such as: listed companies releasing major financial reports, funds publishing quarterly holding reports, securities experiencing abnormal fluctuations and being suspended from trading, regulatory authorities issuing important policy adjustments, and major market risk events (such as a single-day drop in the stock index exceeding 5%). These triggers are monitored in real-time through market dynamics and information disclosure channels. When update conditions are met, the risk characteristic database update process is executed. First, the scope of data to be updated is determined. For timed updates, all historical data from the previous calendar month is collected. For event-triggered updates, the data collection time window is determined based on the scope of the event's impact. For example, when a listed company releases its financial report, all data from the past three months for that security is collected. During the data collection process, an incremental collection strategy can be adopted, acquiring only data that has changed since the last update to improve update efficiency.

[0087] After acquiring the latest original information, the data is reprocessed, features are extracted from new data, and all affected risk characteristic values ​​are updated according to predetermined calculation methods. For example, when a security releases a new financial report, the risk characteristics related to all corresponding financial indicators are recalculated; when a fund discloses its latest holdings, features such as the fund's holding concentration are recalculated. For sentiment-related features, a sliding time window mechanism is established, retaining only the most recent three months of sentiment data to ensure that the risk score reflects the latest market sentiment. After data recalculation, a database update operation is performed. A transaction processing mechanism is adopted to ensure the atomicity and consistency of data updates. Two update strategies are used: full update and incremental update. For daily fluctuations in data such as fund net asset value and securities trading prices, incremental updates are used; for structured data such as financial reports and fund holdings, version management is adopted, retaining historical versions while updating the current valid version. During the update process, a data rollback mechanism can also be established to quickly restore to the previous valid version when data anomalies are detected. After the update is completed, data quality verification is performed. By comparing the changes in data distribution before and after the update, abnormal update situations are identified. For example, when the risk value of a security fluctuates sharply in a short period, a manual review process is triggered. Simultaneously, the integrity and consistency of the database are checked to ensure all relationships are correct. Finally, an update log is recorded, including update time, scope, and data version information. This systematic update mechanism ensures that the risk characteristic database always reflects the latest market conditions and risk levels, providing accurate and timely data support for risk monitoring.

[0088] In one or more possible embodiments, this application also provides a risk prediction system, specifically as follows: Figure 3 As shown, it mainly includes: a fund risk monitoring module and a securities risk monitoring module; the fund risk monitoring module includes:

[0089] Performance Anomaly Monitoring Unit: Captures and stores the performance data of public funds in real time, including key indicators such as net asset value growth rate, return volatility, and Sharpe ratio; by comparing and analyzing with historical data, similar funds, and market benchmarks, it identifies abnormal fluctuations in funds and provides data support for risk warning.

[0090] Trading Restriction Execution Unit: Based on preset trading rules and restrictions (such as position ratio, trading frequency, etc.), it monitors the fund's trading behavior in real time; once illegal trading or abnormal trading patterns are detected, it immediately triggers an early warning mechanism to ensure the compliance and stability of the fund's operation.

[0091] Public opinion monitoring unit: Utilizes web crawling and natural language processing technologies to capture and analyze publicly available information such as news, forums, and social media on the internet in real time; assesses the positive or negative tendencies of information through sentiment analysis algorithms, and extracts risk signals related to the fund, such as negative reports and changes in investor sentiment.

[0092] Comprehensive Risk Monitoring Unit: Combining the data from the above units, the system uses a risk control model to conduct a comprehensive assessment; it identifies potential market risks, credit risks, and liquidity risks, and sets early warning thresholds based on risk levels; when the risk value exceeds the threshold, the system automatically triggers an early warning to remind the user to take appropriate measures.

[0093] The securities risk monitoring module includes:

[0094] Trading Anomaly Monitoring Unit: Monitors securities trading behavior in real time, including key indicators such as trading volume, trading price, and bid-ask spread; identifies abnormal trading patterns, such as unusually large transactions and frequent transactions, by comparing and analyzing historical data and market benchmarks, providing data support for securities risk early warning.

[0095] Financial Analysis Unit: Focuses on assessing the profitability, solvency, growth potential, and cash flow quality of listed companies, identifying issues such as profit volatility, excessive leverage, or tight cash flow, and providing quantitative basis for judging the overall risk exposure of the portfolio and adjusting holdings.

[0096] Public opinion risk monitoring unit: Utilizing web crawling and natural language processing technologies, it captures and analyzes publicly available information from the internet, such as news, forums, and social media, in real time to extract negative information related to securities and changes in investor sentiment.

[0097] Valuation Analysis Unit: Utilizing quantitative valuation models and combining multi-dimensional data such as macroeconomic indicators, market sentiment, historical performance, and industry trends, this unit predicts the reasonable valuation of securities. When the actual valuation deviates from the predicted value to a preset threshold, it identifies the specific valuation situation and provides data support for risk warning.

[0098] Comprehensive Risk Monitoring Unit: Combining the data from the above units, a comprehensive assessment is conducted using a risk control model to identify potential risks and set early warning thresholds based on the risk level.

[0099] Corresponding to the above-mentioned risk prediction method, the present invention also proposes a risk pre-setting device, specifically as follows: Figure 4 As shown, it includes:

[0100] The direct risk value determination module 401 is used to obtain the fund identifier and fund characteristics of the target fund, and match the fund characteristics with a pre-built risk characteristic database according to the fund identifier to obtain the direct risk value corresponding to the target fund;

[0101] The indirect risk value determination module 402 is used to obtain the corresponding security identifier and security characteristics based on the target fund's holding information, and match the security characteristics with a pre-built risk characteristic database based on the security identifier to obtain the indirect risk value corresponding to the target fund's holding information.

[0102] The early warning module 403 determines the comprehensive risk value corresponding to the target fund based on the direct risk value and the indirect risk value, and determines whether to issue an early warning to the target fund based on the comprehensive risk value.

[0103] Since the device embodiments of the present invention correspond to the method embodiments described above, details not disclosed in the device embodiments can be referred to in the method embodiments described above, and will not be repeated in the present invention.

[0104] This application also provides an electronic device, including at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the risk prediction method described above.

[0105] like Figure 5 As shown, the device includes a processor 501, a memory 502, a communication interface 503, and a bus 504. The processor 501, memory 502, and communication interface 503 are interconnected via the bus 504.

[0106] Processor 501 is configured to read instructions from memory 502 and execute them, so that at least one processor can perform the risk prediction method provided in the above embodiments.

[0107] The memory 502 is used to store various instructions and programs for the risk prediction method provided in the above embodiments.

[0108] Bus 504 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0109] Processor 501 can be a central processing unit (CPU), a network processor (NP), a graphics processing unit (GPU), or any combination of CPU, NP, and GPU. It can also be a hardware chip. The aforementioned hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The aforementioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0110] In addition, this application also provides a computer-readable storage medium, such as Figure 6 As shown, the computer storage medium stores a computer program that is used to cause the computer to perform any of the methods described in the above embodiments.

[0111] The memory may include a readable medium in the form of volatile memory, such as random access memory (RAM) 601 and / or cache memory 602, and may further include read-only memory (ROM) 603.

[0112] The memory may also include a program / utility 605 having a set (at least one) of program modules 604, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0113] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0114] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0115] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0116] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0117] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A risk prediction method, characterized by, The method comprises the following steps: obtaining the fund identification and fund characteristics of the target fund, matching the fund characteristics with a pre-constructed risk characteristic database according to the fund identification, and obtaining the direct risk value corresponding to the target fund; obtaining the security identification and security characteristics corresponding to the holding information of the target fund, matching the security characteristics with the pre-constructed risk characteristic database according to the security identification, and obtaining the indirect risk value corresponding to the holding information of the target fund; determining the comprehensive risk value corresponding to the target fund according to the direct risk value and the indirect risk value, and determining whether to issue a warning to the target fund according to the comprehensive risk value.

2. The method of claim 1, wherein, The risk characteristic database is constructed in the following manner: collecting the original information of funds and securities, which includes fund names, fund net values, fund financial reports, security transaction information, security corresponding company financial statements, news information, social media public opinions, and macro market data; from the original information, obtaining the risk characteristics corresponding to each fund and security respectively; the risk characteristics include at least one of the following: fund type, net value volatility, Sharpe ratio, maximum drawdown, security trading volume, security trading price, financial indicators, negative public opinion degree, and valuation deviation; determining the risk value corresponding to each risk characteristic based on a pre-set risk value determination method; associating the identification of funds and securities, the corresponding risk characteristics, and the corresponding risk values to form a risk characteristic database; after determining that the construction of the risk characteristic database is completed, further comprising: determining whether to update the risk characteristic database when the risk characteristic database update condition is met.

3. The method of claim 2, wherein, The negative public opinion degree is obtained in the following manner: using natural language processing technology to perform semantic analysis on the collected news information and social media public opinions to obtain structured data about each fund and security; based on a sentiment analysis algorithm, analyzing the structured data to determine the negative public opinion degree corresponding to each fund and security.

4. The method of claim 1, wherein, Determining whether to issue a warning to the target fund according to the comprehensive risk value comprises: comparing the comprehensive risk value with a plurality of pre-set risk level thresholds; determining the risk level of the target fund according to the comparison result; when the risk level reaches or exceeds a pre-set warning level, generating and outputting warning information containing the risk level.

5. The method of claim 1, wherein, The method comprises the following steps: obtaining the security identification and security characteristics corresponding to the holding information of the target fund, matching the security characteristics with the pre-constructed risk characteristic database according to the security identification, and obtaining the indirect risk value corresponding to the holding information of the target fund; obtaining all securities held by the target fund and the holding proportion of each security according to the holding information of the target fund; matching the security characteristics of each security with the risk characteristic database to obtain the individual risk value of each security; 6. The method of claim 1, wherein, weighting and aggregating the individual risk values according to the holding proportions of the securities to obtain the indirect risk value. The method further comprises:

7. A risk prediction apparatus, characterized by, obtaining the transaction behavior of the target fund, and issuing a warning to the target fund when the transaction behavior exceeds a pre-set transaction rule. The method comprises the following steps: The direct risk value determination module is configured to acquire a fund identifier and fund characteristics of a target fund, match the fund characteristics with a pre-constructed risk characteristic database according to the fund identifier, and acquire a direct risk value corresponding to the target fund. The indirect risk value determination module is configured to acquire a security identifier and security characteristics according to the holding information of the target fund, match the security characteristics with the pre-constructed risk characteristic database according to the security identifier, and acquire an indirect risk value corresponding to the holding information of the target fund. The early warning module is configured to determine a comprehensive risk value corresponding to the target fund according to the direct risk value and the indirect risk value, and determine whether to issue a warning to the target fund according to the comprehensive risk value.

8. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any one of the methods of claims 1-6.

9. A computer storage medium, characterized in that The computer storage medium stores a computer program, and the computer program is used to enable a computer to perform any one of the methods of claims 1-6.

Images