Product risk assessment method and device, program product and storage medium
By determining multiple risk dimensions and their correlation degree of financial products, and using the target scoring network to calculate risk index scores, the problem of inaccurate risk assessment in the existing technology is solved, and a comprehensive assessment and risk management of investment risks of financial products is achieved. .
Patent Information
- Application Number
- CN202411116345.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-12-06
AI Technical Summary
The existing financial product risk research methods focus on the analysis of a single risk dimension, and cannot effectively consider the relationship between multiple risk dimensions, resulting in the inaccurate assessment of investment risk and the inability to control risks in advance.
By determining multiple risk dimensions of the product to be evaluated, relevant market environment information and trading behavior data are obtained, the regression analysis model is used to determine the degree of correlation between the risk dimensions, and the risk indicator score is calculated through the target scoring network, and the investment risk is finally generated grade.
It has achieved an accurate assessment of the investment risks of financial products, and can fully understand the impact of multiple risk dimensions, help investors formulate effective risk management strategies and reduce investment risks.
Smart Images

Figure CN119090647A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of computers, and specifically, to a product risk assessment method and device, a program product, and a storage medium. Background Art
[0002] In recent years, domestic and foreign financial markets have experienced drastic fluctuations, and international market risk events have occurred frequently, bringing greater risks to investors. The risk characteristics and changing patterns of financial products in the financial market are of great significance to the risk management of investors and financial institutions.
[0003] Existing product risk research methods mostly focus on the analysis of a single risk dimension, or although they consider multiple risk dimensions, they lack in-depth research and quantitative analysis of the relationship between these dimensions. This leads to inaccurate assessment of product investment risks and inability to effectively conduct ex ante risk control. Summary of the invention
[0004] The embodiments of the present application provide a product risk assessment method and device, a program product, and a storage medium to at least solve the problem in the related art that product investment risks cannot be accurately assessed.
[0005] According to an embodiment of the present application, a method for risk research on domestic commodity futures varieties is provided, comprising: determining N risk dimensions of a product to be evaluated, wherein the risk dimensions include market environment information related to the product to be evaluated, each risk dimension includes one or more risk indicators, and N is a natural number greater than 1; obtaining indicator data of the N risk dimensions to obtain N groups of indicator data, wherein a group of the indicator data corresponds to risk data of one or more risk indicators under one risk dimension, and the risk data at least includes trading behavior data of the product to be evaluated; determining N target indicator scores based on the N groups of the indicator data, wherein the target indicator scores are used to represent the product performance of the product to be evaluated under the risk dimensions; and determining the risk assessment result of the product to be evaluated using the N target indicator scores, wherein the risk assessment result is used to represent the investment risk level of the product to be evaluated.
[0006] In an exemplary embodiment, determining N risk dimensions of a product to be evaluated includes: obtaining M groups of product data of the product to be evaluated, wherein one group of the product data corresponds to one risk dimension, the product data includes transaction behavior data of the product to be evaluated in a target trading platform, marketization data of the product to be evaluated in a target trading market, industrial chain data of the product to be evaluated, and policy data related to the product to be evaluated, and M is a natural number greater than or equal to N; inputting the M groups of product data into a regression analysis model to obtain the degree of correlation between the product to be evaluated and the multiple risk dimensions output by the regression analysis model; sorting the multiple risk dimensions according to the degree of correlation between the product to be evaluated and the multiple risk dimensions, and determining N risk dimensions from the sorting.
[0007] In an exemplary embodiment, obtaining indicator data of N of the above-mentioned risk dimensions to obtain N groups of indicator data includes: obtaining indicator data of each of the above-mentioned risk dimensions through the following steps to obtain N groups of the above-mentioned indicator data: selecting one or more of the above-mentioned risk indicators under the above-mentioned risk dimensions according to the type of the above-mentioned risk dimensions; obtaining the above-mentioned risk data from the target trading platform and / or the target trading market where the above-mentioned product to be evaluated is located according to the above-mentioned risk indicators, wherein the above-mentioned risk data includes the above-mentioned transaction behavior data, market risk data, industrial chain risk data and policy risk data of the above-mentioned product to be evaluated; and determining the above-mentioned risk data corresponding to the above-mentioned risk indicators as the above-mentioned indicator data.
[0008] In an exemplary embodiment, N target indicator scores are determined based on N groups of the above indicator data, including: calculating the indicator score of each group of the above indicator data through the following steps to obtain N groups of the above indicator scores: inputting the above indicator data into a pre-trained target scoring network to obtain the above indicator scores output by the above target scoring network, wherein the above target scoring network includes: an indicator conversion network, an indicator feature extraction network and an indicator classification network, for each of the above indicator data in each group of the above indicator data, the above indicator conversion network is used to convert the above indicator data into a preset standard value within a preset numerical range to obtain a numerical matrix, the above indicator feature extraction network is used to extract the characteristic vector corresponding to the product risk factor in the above indicator data from the above numerical matrix, and the above indicator classification network is used to assign the indicator score of the above indicator data based on the characteristic vector of the above product risk factor; based on the N groups of the above indicator scores, N target indicator scores are determined.
[0009] In an exemplary embodiment, N of the above-mentioned target indicator scores are determined based on N groups of the above-mentioned indicator scores, including: calculating the target indicator score corresponding to each group of the above-mentioned indicator scores through the following steps to obtain the N groups of the above-mentioned target indicator scores: calculating the weighted indicator score of each of the above-mentioned indicator scores according to the weight of each of the above-mentioned indicator scores in a group of the above-mentioned indicator scores to obtain a group of the above-mentioned weighted indicator scores, wherein the weight of each of the above-mentioned indicator scores is determined based on the risk impact of the risk factors in the corresponding indicator data on the above-mentioned product to be evaluated; calculating the average value of a group of the above-mentioned weighted indicator scores to obtain the above-mentioned target indicator score.
[0010] In an exemplary embodiment, the risk assessment result of the product to be assessed is determined using N of the target indicator scores, including: screening scores greater than the preset indicator scores from the N target indicator scores to obtain K of the target indicator scores, wherein the K is a natural number less than or equal to the N; when K is greater than or equal to a first preset number, generating the risk assessment result, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is greater than or equal to the preset risk level; when K is less than the first preset number, generating the risk assessment result, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is less than the preset risk level; wherein the K target indicator scores correspond to K risk dimensions, each of the K risk dimensions includes a second preset number of risk indicators, and the risk level of the second preset number of risk indicators is greater than the preset risk level.
[0011] According to another embodiment of the present application, a risk assessment device for a product is provided, comprising: a first determination module, used to determine N risk dimensions of the product to be assessed, wherein the risk dimensions include market environment information related to the product to be assessed, each of the risk dimensions includes one or more risk indicators, and N is a natural number greater than 1; a first acquisition module, used to acquire indicator data of the N risk dimensions to obtain N groups of indicator data, wherein one group of the indicator data corresponds to risk data of one or more risk indicators under one risk dimension, and the risk data at least includes transaction behavior data of the product to be assessed; a second determination module, used to determine N target indicator scores based on the N groups of indicator data, wherein the target indicator scores are used to represent the product performance of the product to be assessed under the risk dimensions; a third determination module, used to determine the risk assessment result of the product to be assessed using the N target indicator scores, wherein the risk assessment result is used to represent the investment risk level of the product to be assessed.
[0012] According to another embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.
[0013] According to another embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.
[0014] According to another embodiment of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in any one of the above method embodiments when executing the computer program.
[0015] Through the present application, N risk dimensions of the product to be evaluated are determined, indicator data of the N risk dimensions are obtained, and a target indicator score for representing the product performance of the product to be evaluated under the risk dimension is determined based on the indicator data. The target indicator score is then used to determine the investment risk level of the product to be evaluated. Therefore, the problem that the investment risk of a product cannot be accurately evaluated in the related technology can be solved, and the effect of accurately evaluating the investment risk of a product can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 is a schematic diagram of a hardware environment of a product risk assessment method according to an embodiment of the present application;
[0017] Figure 2 is a flow chart of a product risk assessment method according to an embodiment of the present application;
[0018] Figure 3 is a schematic diagram of risk assessment of a target futures product according to an embodiment of the present application;
[0019] Figure 4 is a schematic diagram of a risk assessment index system for a target futures product according to an embodiment of the present application;
[0020] Figure 5 is a schematic diagram of the results of risk assessment of a target futures product according to an embodiment of the present application;
[0021] Figure 6 It is a structural block diagram of a risk assessment device for a product according to an embodiment of the present application. DETAILED DESCRIPTION
[0022] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.
[0024] The method embodiments provided in the embodiments of the present application can be executed in a server device or a similar computing device. Taking running on a server device as an example, Figure 1 FIG. 1 is a schematic diagram of the hardware environment of a risk assessment method for a product in an embodiment of the present application. Figure 1 As shown, the server device may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the server device may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above server device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.
[0025] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the risk assessment method of the product in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the server device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0026] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the server device. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.
[0027] In this embodiment, a product risk assessment method is provided. Figure 2 is a flow chart of a method for risk assessment of a product according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:
[0028] Step S202, determining N risk dimensions of the product to be evaluated, wherein the risk dimensions include market environment information related to the product to be evaluated, each of the risk dimensions includes one or more risk indicators, and N is a natural number greater than 1;
[0029] Step S204, obtaining indicator data of N risk dimensions to obtain N groups of indicator data, wherein one group of indicator data corresponds to risk data of one or more risk indicators under one risk dimension, and the risk data at least includes transaction behavior data of the product to be evaluated;
[0030] Step S206, determining N target indicator scores based on the N groups of indicator data, wherein the target indicator scores are used to represent the product performance of the product to be evaluated under the risk dimension;
[0031] Step S208, using the N target indicator scores to determine the risk assessment result of the product to be assessed, wherein the risk assessment result is used to represent the investment risk level of the product to be assessed.
[0032] This embodiment includes but is not limited to being applied to scenarios of risk assessment of investment products in the financial market, such as scenarios of assessing the investment risks of various futures in the futures market and scenarios of assessing the investment risks of stocks. By assessing the price volatility risk, liquidity risk, leverage risk, etc. of investment products, it can help investors better formulate risk management strategies, including setting stop-loss points, position control, etc., thereby reducing investment risks.
[0033] Optionally, in the process of risk assessment, this embodiment needs to consider multiple aspects related to the product to be assessed to ensure a comprehensive understanding of the possible risks. Risk dimensions include but are not limited to the transaction status, market status, industrial chain status, policy status, and interaction between users and trading platforms of the product to be assessed. For example, in the case where the product to be assessed is futures, the risk dimensions of futures include: Figure 3 As shown, there are four risk dimensions.
[0034] Optionally, the risk indicators in this embodiment include but are not limited to the classification under each risk dimension. For example, the risk indicators included in the transaction situation include trading volume, open interest, delivery rate, etc. The interaction between the user and the trading platform includes the time and number of times used to log in to the trading platform, the time and number of times to browse the product to be evaluated, etc. For example, in the case where the product to be evaluated is futures, the risk indicators of futures include but are not limited to the trading volume, open interest, delivery rate, delivery volume, delivery product requirements, etc. under the basic conditions of industry and spot. Price fluctuations in the futures market, operational risks caused by errors in the trading platform, price trends, fluctuations, trading volumes and other data of futures varieties can help analyze market trends; information such as the supply and demand situation, output, inventory, policies and regulations of related commodities can help understand the fundamental factors behind the market; various technical indicators, chart patterns and other analytical tools can help analyze market trends; domestic and foreign economic data, interest rates, exchange rates and other information can help understand the impact of the macroeconomic environment on the market; domestic and foreign events, geopolitical risks and other information can help understand the uncertainty factors in the market. For example, the risk situation of commodity futures varieties is mainly affected by factors such as futures trading conditions, market conditions, industrial chain conditions, and policy conditions. Optionally, the target index score in this embodiment can be comprehensively determined based on various factors such as historical data, industry comparison, expert evaluation, etc. For example, the trading volume index score is comprehensively determined based on various factors such as historical trading volume data, industry comparison with other exchanges, and expert evaluation; the volatility index score is comprehensively determined based on factors such as historical volatility data, industry comparison, and expert evaluation; and the profit index score is comprehensively determined based on factors such as historical profit data, industry comparison, and expert evaluation.
[0035] Optionally, the risk assessment results in this embodiment are presented in a form that can intuitively display the results, for example, an evaluation result form including various indicator scores of the target product and the risk level of the final product; or an evaluation result matrix formed by arranging the indicator data and its corresponding indicator scores on the horizontal and vertical axes.
[0036] Through the above steps, by determining N risk dimensions of the product to be evaluated, obtaining indicator data of the risk dimension and determining the target indicator score for representing the product performance of the product to be evaluated under the risk dimension based on the indicator data, and then using the target indicator score to determine the risk assessment result of the product to be evaluated, the problem that the investment risk of the product cannot be accurately assessed in the relevant technology can be solved, and the effect of accurately assessing the investment risk of the product can be achieved.
[0037] In an exemplary embodiment, determining N risk dimensions of a product to be evaluated includes: obtaining M groups of product data of the product to be evaluated, wherein one group of product data corresponds to one risk dimension, the product data includes transaction behavior data of the product to be evaluated in a target trading platform, marketization data of the product to be evaluated in a target trading market, industrial chain data of the product to be evaluated, and policy data related to the product to be evaluated, and M is a natural number greater than or equal to N; inputting the M groups of product data into a regression analysis model to obtain the degree of correlation between the product to be evaluated and the multiple risk dimensions output by the regression analysis model; sorting the multiple risk dimensions according to the degree of correlation between the product to be evaluated and the multiple risk dimensions, and determining N risk dimensions from the sorting.
[0038] Optionally, in this embodiment, the M group of product data of the product to be evaluated can be obtained in the following ways: For example, obtaining data through the official website where the target product is traded; obtaining data using the platform of a professional financial data service provider; obtaining and analyzing data by using trading software or data analysis tools.
[0039] Among them, the risk dimensions include but are not limited to the risk dimensions corresponding to the product data described in the above embodiments, which will not be repeated here.
[0040] Optionally, in this embodiment, the regression analysis model can be trained in the following manner: collect historical data of multiple groups of different products, including product features, sales data, customer feedback, market environment, user interaction data with the trading platform (for example, page views, page views, purchases, purchase time, purchased products), etc. Extract risk-related features from historical data. For example, these features include product yield, volatility, liquidity, credit rating, etc. Determine the risk dimensions that need to be evaluated, such as market risk, credit risk, liquidity risk, etc. Select a suitable regression model based on data characteristics and problem requirements. For example, linear regression, ridge regression, LASSO regression, elastic network regression, etc. Clean the historical data and handle missing values, outliers, etc. And standardize or normalize the data to eliminate the influence of different dimensions. Use the cleaned historical data set to train the regression model. Divide the data set into a training set and a test set to avoid overfitting. Use the test set to evaluate the performance of the model. Analyze the model coefficients to explain the degree of influence of different features on the risk dimension. A positive coefficient indicates that the feature is positively correlated with the risk, and a negative coefficient indicates a negative correlation. The trained regression model can be used to effectively evaluate the correlation between the product to be evaluated and multiple risk dimensions from multiple sets of product data. This helps to better understand product risks and make more informed investment decisions.
[0041] Optionally, in the process of multi-dimensional risk assessment, the purpose of sorting multiple risk dimensions in this embodiment is to determine which risk dimensions have a higher impact and importance on the product to be assessed. This embodiment can sort the risk dimensions in the following manner:
[0042] S1, assign a weight to each risk dimension to reflect its importance in the assessment process. The weight can be determined based on historical data, expert opinion, or other factors.
[0043] S2, for each risk dimension, use a scorecard or other assessment tool to calculate a risk value. The risk value can be a probability, loss amount, or other risk-related measure.
[0044] S3, multiply the risk value of each risk dimension by its weight to obtain the weighted risk value.
[0045] S4, sort the risk dimensions in descending order according to the weighted risk value. The risk dimension with the highest weighted risk value will be ranked first.
[0046] S5, select the top N risk dimensions from the ranking, which will be considered as the risk dimensions with the highest impact and importance to the product to be evaluated.
[0047] For example, suppose there is a financial product to be evaluated, and the following five risk dimensions need to be evaluated: credit risk, market risk, operational risk, compliance risk, and liquidity risk. Assign weights to each risk dimension and use a scorecard to calculate the risk value. The following is the calculation process of the weighted risk value:
[0048] Credit risk: weight 0.3, risk value 0.8, weighted risk value 0.24; market risk: weight 0.2, risk value 0.7, weighted risk value 0.14; operational risk: weight 0.15, risk value 0.9, weighted risk value 0.135; compliance risk: weight 0.2, risk value 0.6, weighted risk value 0.12; liquidity risk: weight 0.15, risk value 0.5, weighted risk value 0.075. Sorting the risk dimensions according to the weighted risk value yields the following order: credit risk > operational risk > market risk > compliance risk > liquidity risk. Assuming that the first three risk dimensions are selected, credit risk, operational risk and market risk will be considered as the risk dimensions with the highest impact and importance on financial products. Through the above steps, the product data of the product to be evaluated is input into the regression analysis model to obtain the degree of correlation between the product to be evaluated and the multiple risk dimensions, and further determine the risk dimensions. It can solve the problem that risk assessment in related technologies is highly subjective, resulting in inaccurate risk assessment results, and achieve the effect of comprehensively assessing product risks.
[0049] In an exemplary embodiment, obtaining indicator data of N of the above-mentioned risk dimensions to obtain N groups of indicator data includes: obtaining indicator data of each of the above-mentioned risk dimensions through the following steps to obtain N groups of the above-mentioned indicator data: selecting one or more of the above-mentioned risk indicators under the above-mentioned risk dimensions according to the type of the above-mentioned risk dimensions; obtaining the above-mentioned risk data from the target trading platform and / or the target trading market where the above-mentioned product to be evaluated is located according to the above-mentioned risk indicators, wherein the above-mentioned risk data includes the above-mentioned transaction behavior data, market risk data, industrial chain risk data and policy risk data of the above-mentioned product to be evaluated; and determining the above-mentioned risk data corresponding to the above-mentioned risk indicators as the above-mentioned indicator data.
[0050] Optionally, in this embodiment, the method of selecting one or more risk indicators under the risk dimension according to the type of risk dimension may include: presetting the risk indicators included in each risk dimension of the target product, displaying the risk indicators in the form of a table, and further determining the appropriate risk indicators by screening the risk dimensions in the table; or inputting the risk indicators included in each risk dimension of the pre-set target product into a program code, and executing the program to output the appropriate risk indicators by inputting the type of risk dimension.
[0051] Optionally, this embodiment may select a target trading platform to obtain risk data based on the number of visits, visit frequency, and visit time of multiple trading platforms visited by the user.
[0052] For example, collect data on users' visits to various trading platforms in the past period of time, including the number of visits, frequency of visits, and time of visits. Remove outliers or erroneous data. Analyze the user's access behavior on each platform, including the size of the number of visits, the frequency of visits, and the distribution of the time of visits. Evaluate the popularity and user activity of each trading platform based on the user's behavior pattern. Analyze the historical risk events of each trading platform, such as security vulnerabilities, fraudulent behavior, market manipulation, etc. Build a risk assessment model to combine user access behavior with the historical risk events of the trading platform to assess the risk level of each platform. According to the results of the risk assessment model, rank the risks of each trading platform. Select the platform with the highest risk in the risk assessment results as the target trading platform in order to obtain more comprehensive risk data. Obtain relevant risk data from the target trading platform, such as transaction data, user behavior data, security event records, etc. Integrate the acquired risk data and analyze its possible impact on users and the entire market. In the above way, a target trading platform can be scientifically selected, its risk data can be obtained, and support can be provided for further risk management and decision-making.
[0053] This embodiment can also access the risk data in the target trading platform through the application programming interface of the target trading platform under the user's trigger instruction, including transaction data, user behavior data, historical price data, etc. It should be noted that this embodiment can also monitor the target trading platform in real time to obtain real-time data generated in the target trading platform in real time.
[0054] Through the above steps, by selecting risk indicators under the risk dimension according to the type of risk dimension, obtaining risk data based on the risk indicator and determining it as indicator data, it can include various types of indicator data, thereby solving the problem that the risk assessment method in the prior art is relatively single and difficult to fully reflect the risk characteristics of the product.
[0055] In an exemplary embodiment, N target indicator scores are determined based on N groups of the above indicator data, including: calculating the indicator score of each group of the above indicator data through the following steps to obtain N groups of the above indicator scores: inputting the above indicator data into a pre-trained target scoring network to obtain the above indicator scores output by the above target scoring network, wherein the above target scoring network includes: an indicator conversion network, an indicator feature extraction network and an indicator classification network, for each of the above indicator data in each group of the above indicator data, the above indicator conversion network is used to convert the above indicator data into a preset standard value within a preset numerical range to obtain a numerical matrix, the above indicator feature extraction network is used to extract the characteristic vector corresponding to the product risk factor in the above indicator data from the above numerical matrix, and the above indicator classification network is used to assign the indicator score of the above indicator data based on the characteristic vector of the above product risk factor; based on the N groups of the above indicator scores, N target indicator scores are determined.
[0056] Optionally, the methods of training the target scoring network in this embodiment include but are not limited to: using a labeled data set to train the target scoring network, allowing the network to learn how to predict the target score based on the input data for supervised learning; using an unlabeled data set to train the target scoring network, allowing the network to learn the patterns and relationships between the data by itself, thereby performing unsupervised learning for target score prediction; combining labeled and unlabeled data to train the target scoring network to improve the performance and generalization ability of the network for semi-supervised learning; using reinforcement learning to train the target scoring network, optimizing the network's behavior and decision-making strategy through interaction and feedback with the environment. It should be noted that the above methods can be used alone or in combination to improve the performance and effect of the target scoring network.
[0057] Optionally, before the indicator conversion network in this embodiment converts the indicator data into a preset standard value within a preset value range, it will first pre-process the input indicator data, including checking whether the indicator data is reasonable and whether there are missing indicator values, performing alarm processing for the existence of missing indicator data, determining the content of the missing indicator data; performing cleaning operations on unreasonable data, etc.
[0058] Optionally, the indicator conversion network in this embodiment converts the indicator data into preset standard values within a preset numerical range, and the steps of obtaining a numerical matrix include: preprocessing the input indicator data; inputting the preprocessed indicator data into a trained indicator conversion network; scoring the indicator data according to a preset numerical range (for example, from 1 to 5, where 1 represents the lowest and 5 represents the highest) to obtain a numerical matrix.
[0059] Optionally, the method of allocating indicator scores of indicator data based on the characteristic vector of product risk factors in this embodiment includes but is not limited to: clarifying the degree of influence of various product risk factors on the risk of the product according to the characteristic vector of the product risk factor, and further allocating indicator scores of the indicator data according to the degree of influence.
[0060] Through the above steps, the indicator data is input into the pre-trained target scoring network to obtain the indicator score, and the corresponding target indicator score is determined based on the indicator score, which simplifies the process of scoring the target indicator and improves the efficiency of scoring the indicator data.
[0061] In an exemplary embodiment, N of the above-mentioned target indicator scores are determined based on N groups of the above-mentioned indicator scores, including: calculating the target indicator score corresponding to each group of the above-mentioned indicator scores through the following steps to obtain the N groups of the above-mentioned target indicator scores: calculating the weighted indicator score of each of the above-mentioned indicator scores according to the weight of each of the above-mentioned indicator scores in a group of the above-mentioned indicator scores to obtain a group of the above-mentioned weighted indicator scores, wherein the weight of each of the above-mentioned indicator scores is determined based on the risk impact of the risk factors in the corresponding indicator data on the above-mentioned product to be evaluated; calculating the average value of a group of the above-mentioned weighted indicator scores to obtain the above-mentioned target indicator score.
[0062] Optionally, the weight of the indicator score in this embodiment is represented by: directly giving the numerical value of the data to represent the weight, the larger the numerical value, the higher the weight (or the smaller the numerical value, the higher the weight); converting the data into a percentage to represent the weight, for example, normalizing the data to a percentage value of 0-100; representing the weight of the data in proportion, for example, the weight of A is twice that of B; grading the data to represent the weight, for example, dividing the data into three levels: high, medium, and low, each corresponding to a different weight.
[0063] Optionally, the display form of the weight of the indicator score in this embodiment includes but is not limited to: bar charts, pie charts, radar charts and other charts to display the proportion and relationship of data weights; using visualization tools such as word clouds, heat maps, etc. to display the distribution of data weights; explaining the meaning of data weights through text descriptions and explanations.
[0064] Through the above steps, the target indicator score is obtained by calculating the indicator score according to its corresponding weighted score and calculating the average of the weighted indicator score. The obtained target indicator score is more reasonable, accurate and objective, and the calculated target indicator score improves the comprehensiveness of the product risk assessment results.
[0065] In an exemplary embodiment, the risk assessment result of the product to be assessed is determined using N of the target indicator scores, including: screening scores greater than the preset indicator scores from the N target indicator scores to obtain K of the target indicator scores, wherein the K is a natural number less than or equal to the N; when K is greater than or equal to a first preset number, generating the risk assessment result, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is greater than or equal to the preset risk level; when K is less than the first preset number, generating the risk assessment result, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is less than the preset risk level; wherein the K target indicator scores correspond to K risk dimensions, each of the K risk dimensions includes a second preset number of risk indicators, and the risk level of the second preset number of risk indicators is greater than the preset risk level.
[0066] Optionally, the target indicator score is determined comprehensively based on various factors including but not limited to historical data, industry comparison, expert evaluation, etc., which will not be elaborated here.
[0067] Optionally, the risk assessment results are presented in a form that can intuitively display the results, which will not be described in detail here.
[0068] In this embodiment, for example, for each product variety, the respective indicator scores under the four main risk dimensions are calculated respectively. If two or more risk indicators are scored as "high" under a certain risk dimension, then the risk of this risk dimension can be considered to be low. Then, the comprehensive situation of the four main risk dimensions is calculated. If two or more of the main risk dimensions are judged to be low risk (that is, each has two or more high-scoring indicators), then the risk of the entire product is evaluated as low risk. On the contrary, if two or more risk indicators are scored as "low" under a certain main risk dimension, then the risk of this main aspect is high. If two or more of the four main risk dimensions are judged to be high risk (that is, each has two or more low-scoring indicators), then the risk of the entire product is evaluated as high risk. This ensures that when assessing product risks, the various indicators and main aspects can be fully considered, thereby obtaining a relatively accurate risk assessment result.
[0069] Through the above steps, by screening the target indicator scores that are greater than the preset indicator scores, the number of target indicator scores that meet the requirements is obtained, and it is compared with the preset number to generate a risk assessment result. The method of generating risk assessment results is clarified, and the scores and quantities are compared intuitively, so that the generated risk assessment results are more accurate.
[0070] The risk assessment method of the product in the embodiment of the present application is explained below in conjunction with an optional example. In this optional example, the product to be assessed is explained by taking a futures product as an example.
[0071] The risk of futures products in this optional example is mainly affected by factors such as futures trading conditions, market conditions, industrial chain conditions, policy conditions, and the interaction between traders and trading platforms. For example, after collecting information on the above-mentioned influencing factors, determining the risk assessment method for domestic commodity futures products may include the following steps:
[0072] Step 1, obtaining the target object's behavior information from each trading platform, wherein the behavior information includes the target object's browsing or futures trading information through each trading platform;
[0073] Step 2, determine the target futures product to be risk assessed from the target object's behavior information, and extract the target object's assessment needs for the target futures product; and select the target trading platform based on the trading status of the target futures product in each trading platform; for example, the target object logs in to each trading platform every day to browse the trading status of a certain futures product, so it can be determined that the target object has a high degree of attention to the futures product, and thus it can be predicted that the target object has a demand for risk assessment of the futures product; determine the trading platform with the most login times as the target trading platform, or determine the trading platform with a relatively large trading volume as the target trading platform;
[0074] Step 3, obtain the transaction information of the target futures product from the target trading platform, and determine the risk dimension of the target futures product from the transaction information, for example Figure 3 The four risk dimensions shown mainly include futures trading conditions, market conditions, industrial chain conditions and policy conditions.
[0075] Step 4: After the risk dimensions are determined, construct the following risk indicators based on the risk indicators included in each risk dimension: Figure 4 The evaluation index system of risk indicators shown in the figure includes fundamental driving forces, actual demand, and policy factors; rules and basic trading conditions are used to evaluate the transaction status, open interest, delivery volume, delivery rate, inventory status, and domestic and foreign correlation of futures products; maturity is used to evaluate the market acceptance, pricing power, and industrial chain status of futures products.
[0076] Step 5: Score the risk indicators in the risk assessment system. The scoring range can be from "high" to "low", or there is a specific numerical range to indicate the level of risk. For example, a scoring range of 1 to 5 is set, where 1 represents the lowest (or highest risk) and 5 represents the highest (or lowest risk). The score of each indicator can be determined based on a variety of factors such as its historical data, industry comparison, and expert evaluation.
[0077] For example, based on historical data, the correlation between price changes and supply and demand fundamentals is evaluated. If the correlation is high, the score corresponding to the fundamental driving force indicator is high (such as 5 points); if the correlation is low, the score is low (such as 1 point); based on the ratio of actual demand to trading volume, if the ratio is high, the market demand score is high (such as 5 points); if the ratio is low, the score is low (such as 1 point); the transaction situation indicator is scored based on the daily average or three-year minimum trading volume and transaction amount. The higher the daily average / three-year minimum trading volume and transaction amount, the higher the score; the more reasonable the layout of the delivery warehouse and the lower the transportation risk, the higher the score of the corresponding delivery location indicator.
[0078] Step 6: Calculate the score of each risk dimension in the risk indicator system. Average all indicator scores under each risk dimension (or weighted average, with different weights set according to the importance of the indicator) to obtain the total score of the risk dimension.
[0079] For example, set a scoring range from "high" to "low". Under this premise, if two or more risk indicators are scored as "high" under a certain risk dimension, then the risk of this risk dimension can be considered to be low.
[0080] Step 7, after obtaining the total score of each risk dimension, comprehensively judge the risk level of the target futures based on these scores. Set a threshold (such as 3 points). If the total score of a risk dimension is higher than this threshold, the risk of the risk dimension is considered to be low, otherwise the risk is considered to be high. Count the number of risk dimensions with low risk. If the number reaches or exceeds 2, the risk of the entire commodity variety is considered to be low, otherwise the risk is considered to be high.
[0081] For example, when the scoring range is set from "high" to "low", and the threshold is set to low risk, if two or more risk dimensions are judged to be low risk (that is, each has two or more high-scoring indicators), then the risk of the entire target futures will be evaluated as low risk; conversely, if under a certain risk dimension, two or more risk indicators are scored as "low", then the risk of this risk dimension is high.
[0082] For example, when the scoring range is set from "high" to "low", the threshold is set as high risk. If two or more of the four risk dimensions are judged to be high risk (i.e., each has two or more low-scoring indicators), then the risk of the target futures is evaluated as high risk.
[0083] Step 8: Based on the risk level of the target futures obtained above, Figure 5 The risk levels corresponding to different target futures are shown in a table.
[0084] Through this optional example, the technical problem of inaccurate assessment of the investment risk level of the product in the relevant technology is solved, so that for specific target futures, its risk can be comprehensively and dynamically assessed, and effective ex ante risk control can be carried out.
[0085] It should be noted that, through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.
[0086] In this embodiment, a product risk assessment device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.
[0087] Figure 6 is a structural block diagram of a risk assessment device for a product according to an embodiment of the present application, such as Figure 6 As shown, the device comprises:
[0088] A first determination module 62 is used to determine N risk dimensions of the product to be evaluated, wherein the risk dimensions include market environment information related to the product to be evaluated, each of the risk dimensions includes one or more risk indicators, and N is a natural number greater than 1;
[0089] The first acquisition module 64 is used to acquire the indicator data of N of the above risk dimensions to obtain N groups of indicator data, wherein one group of the above indicator data corresponds to the risk data of one or more of the above risk indicators under one of the above risk dimensions, and the above risk data at least includes the transaction behavior data of the above product to be evaluated.
[0090] A second determination module 66 is used to determine N target indicator scores based on the N groups of the indicator data, wherein the target indicator scores are used to represent the product performance of the product to be evaluated under the risk dimension;
[0091] The third determination module 68 is used to determine the risk assessment result of the product to be assessed using the N target indicator scores, wherein the risk assessment result is used to represent the investment risk level of the product to be assessed.
[0092] In an exemplary embodiment, the first determination module 62 includes: a first acquisition unit, used to acquire M groups of product data of the product to be evaluated, wherein one group of the product data corresponds to one of the risk dimensions, and the product data includes transaction behavior data of the product to be evaluated in the target trading platform, marketization data of the product to be evaluated in the target trading market, industrial chain data of the product to be evaluated, and policy data related to the product to be evaluated, and M is a natural number greater than or equal to N; a first input unit, used to input the M groups of product data into a regression analysis model to obtain the degree of correlation between the product to be evaluated and the multiple risk dimensions output by the regression analysis model; a first sorting unit, used to sort the multiple risk dimensions according to the degree of correlation between the product to be evaluated and the multiple risk dimensions; a first determination unit, used to determine N risk dimensions from the sorting.
[0093] In an exemplary embodiment, the first acquisition module 64 includes: a first selection unit, used to select one or more of the risk indicators under the risk dimension according to the type of the risk dimension; a first acquisition unit, used to obtain the risk data from the target trading platform and / or target trading market where the product to be evaluated is located according to the risk indicators, wherein the risk data includes the transaction behavior data, market risk data, industrial chain risk data and policy risk data of the product to be evaluated; and a second determination unit, used to determine the risk data corresponding to the risk indicator as the indicator data.
[0094] In an exemplary embodiment, the second determination module 66 is further used to calculate the indicator score of each group of the indicator data through the following steps to obtain N groups of the indicator scores: input the indicator data into a pre-trained target scoring network to obtain the indicator score output by the target scoring network, wherein the target scoring network includes: an indicator conversion network, an indicator feature extraction network and an indicator classification network, for each indicator data in each group of the indicator data, the indicator conversion network is used to convert the indicator data into a preset standard value within a preset numerical range to obtain a numerical matrix, the indicator feature extraction network is used to extract the characteristic vector corresponding to the product risk factor in the indicator data from the numerical matrix, and the indicator classification network is used to assign the indicator score of the indicator data based on the characteristic vector of the product risk factor; determine N target indicator scores based on the N groups of the indicator scores.
[0095] In an exemplary embodiment, the second determination module 66 is further used to calculate the target indicator score corresponding to each group of the indicator scores through the following steps to obtain N groups of the target indicator scores: according to the weight of each indicator score in a group of the indicator scores, calculate the weighted indicator score of each indicator score to obtain a group of the weighted indicator scores, wherein the weight of each indicator score is determined based on the risk impact of the risk factor in the corresponding indicator data on the product to be evaluated; calculate the average value of a group of the weighted indicator scores to obtain the target indicator score.
[0096] In an exemplary embodiment, the third determination module 68 includes: a first screening unit, used to screen scores greater than the preset indicator scores from the N target indicator scores to obtain K target indicator scores, wherein K is a natural number less than or equal to N; a first generating unit, used to generate the risk assessment result when K is greater than or equal to a first preset number, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is greater than or equal to the preset risk level; a second generating unit, used to generate the risk assessment result when K is less than the first preset number, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is less than the preset risk level; wherein the K target indicator scores correspond to K risk dimensions, each of the K risk dimensions includes a second preset number of risk indicators, and the risk level of the second preset number of risk indicators is greater than the preset risk level.
[0097] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.
[0098] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.
[0099] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.
[0100] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.
[0101] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.
[0102] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.
[0103] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.
[0104] The embodiments of the present application also provide a computer program, which includes computer instructions, which are stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of any one of the above method embodiments.
[0105] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.
[0106] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.
[0107] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.
Claims
1. A product risk assessment method, characterized in that: include: Determine N risk dimensions of the product to be evaluated, wherein the risk dimensions include market environment information related to the product to be evaluated, each of the risk dimensions includes one or more risk indicators, and N is a natural number greater than 1; Obtaining indicator data of N risk dimensions to obtain N groups of indicator data, wherein one group of indicator data corresponds to risk data of one or more risk indicators under one risk dimension, and the risk data at least includes transaction behavior data of the product to be evaluated; Determine N target indicator scores based on the N groups of indicator data, wherein the target indicator scores are used to represent the product performance of the product to be evaluated under the risk dimension; The risk assessment result of the product to be assessed is determined using the N target indicator scores, wherein the risk assessment result is used to represent the investment risk level of the product to be assessed.
2. The method according to claim 1, characterized in that Determine the N risk dimensions of the product to be evaluated, including: Acquire M groups of product data of the product to be evaluated, wherein one group of the product data corresponds to one risk dimension, the product data includes transaction behavior data of the product to be evaluated in a target trading platform, marketization data of the product to be evaluated in a target trading market, industrial chain data of the product to be evaluated, and policy data related to the product to be evaluated, and M is a natural number greater than or equal to N; Inputting the M groups of product data into a regression analysis model to obtain the correlation between the product to be evaluated and the multiple risk dimensions output by the regression analysis model; According to the degree of association between the product to be evaluated and the multiple risk dimensions, the multiple risk dimensions are sorted, and N risk dimensions are determined from the sorting.
3. The method according to claim 1, characterized in that Obtain N indicator data of the risk dimensions to obtain N groups of indicator data, including: Obtain the indicator data of each risk dimension through the following steps to obtain N groups of indicator data: Select one or more risk indicators under the risk dimension according to the type of the risk dimension; Acquire the risk data from the target trading platform and / or target trading market where the product to be evaluated is located according to the risk indicator, wherein the risk data includes the transaction behavior data, market risk data, industrial chain risk data and policy risk data of the product to be evaluated; The risk data corresponding to the risk indicator is determined as the indicator data.
4. The method according to claim 1, characterized in that: Determining N target indicator scores based on the N groups of indicator data includes: Calculate the index score of each group of the index data through the following steps to obtain N groups of index scores: Input the indicator data into a pre-trained target scoring network to obtain the indicator score output by the target scoring network, wherein the target scoring network includes: an indicator conversion network, an indicator feature extraction network, and an indicator classification network. For each indicator data in each group of the indicator data, the indicator conversion network is used to convert the indicator data into a preset standard value within a preset value range to obtain a numerical matrix. The indicator feature extraction network is used to extract a feature vector corresponding to a product risk factor in the indicator data from the numerical matrix. The indicator classification network is used to assign an indicator score to the indicator data based on the feature vector of the product risk factor. Determine N target indicator scores based on N groups of indicator scores.
5. The method according to claim 4, characterized in that Determining N target indicator scores based on the N groups of indicator scores includes: Calculate the target indicator score corresponding to each group of the indicator scores through the following steps to obtain N groups of the target indicator scores: According to the weight of each indicator score in the set of indicator scores, a weighted indicator score of each indicator score is calculated to obtain a set of weighted indicator scores, wherein the weight of each indicator score is determined based on the risk impact of the risk factor in the corresponding indicator data on the product to be evaluated; Calculate the average of a set of weighted indicator scores to obtain the target indicator score.
6. The method according to claim 4, characterized in that Determining the risk assessment result of the product to be assessed using the N target indicator scores includes: Filter the scores greater than the preset index score from the N target index scores to obtain K target index scores, where K is a natural number less than or equal to N; In the case where K is greater than or equal to a first preset number, generating the risk assessment result, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is greater than or equal to a preset risk level; In the case where K is less than a first preset number, generating the risk assessment result, wherein the risk assessment result is used to indicate that the investment risk level of the product to be assessed is less than a preset risk level; Among them, the K target indicator scores correspond to K risk dimensions, each of the K risk dimensions includes a second preset number of risk indicators, and the risk levels of the second preset number of risk indicators are all greater than the preset risk level.
7. A product risk assessment device, characterized in that: include: A first determination module is used to determine N risk dimensions of the product to be evaluated, wherein the risk dimensions include market environment information related to the product to be evaluated, each of the risk dimensions includes one or more risk indicators, and N is a natural number greater than 1; A first acquisition module is used to acquire indicator data of N risk dimensions to obtain N groups of indicator data, wherein one group of indicator data corresponds to risk data of one or more risk indicators under one risk dimension, and the risk data at least includes transaction behavior data of the product to be evaluated; A second determination module is used to determine N target indicator scores based on the N groups of indicator data, wherein the target indicator scores are used to represent the product performance of the product to be evaluated under the risk dimension; The third determination module is used to determine the risk assessment result of the product to be assessed by using the N target indicator scores, wherein the risk assessment result is used to represent the investment risk level of the product to be assessed.
8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 6 when executed by a processor.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method described in any one of claims 1 to 6 are implemented.
