Credit risk early warning method, equipment and storage medium for market entities

Abstract

The embodiments of the present application provide a credit risk early warning method, device and storage medium for market entities, and relate to the field of risk early warning technology. The method includes: determining at least one target early warning indicator that needs to be monitored for the target market entity; based on each target early warning indicator, obtaining indicator data corresponding to the target early warning indicator, and performing risk early warning monitoring on the target market entity based on the adaptively adjusted early warning threshold and indicator data; wherein the early warning threshold is determined based on the relationship between historical indicator data and historical risk data. The embodiments of the present application adaptively adjust the early warning threshold based on the relationship between historical indicator data and risk data, and can timely adjust the early warning threshold to adapt to the changes in the relationship between indicators and risks, thereby improving the accuracy of credit risk early warning for market entities.

Description

Technical Field

[0001] The present application relates to the technical field of risk warning, and in particular to a credit risk warning method, device and storage medium for market entities. Background Art

[0002] By scientifically assessing the risk of corporate violations and breaches of trust, and strengthening spot checks on prominent issues and potential risks based on risk assessment results, we can achieve rational allocation and efficient use of regulatory resources, thereby improving the overall effectiveness of supervision and better promoting high-quality economic and social development.

[0003] At present, credit risk warnings for market entities are mainly concentrated on corporate entities. By obtaining certain key indicators of enterprises for continuous monitoring and issuing warnings based on thresholds, market entities with higher credit risks can be discovered in a timely manner.

[0004] However, existing credit risk early warning schemes often directly use the initial threshold set based on business experience for early warning, or fine-tune the initial threshold based on the proportion of objects exceeding the initial threshold to determine the final early warning threshold. The early warning thresholds set by these schemes are relatively fixed and highly dependent on business experience. Therefore, they cannot adapt to changes in risk situations in a timely manner, resulting in low accuracy in credit risk early warnings for market entities. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a credit risk early warning method, device and storage medium for market entities, so as to improve the accuracy of credit risk early warning for market entities.

[0006] In a first aspect, the present application embodiment provides a credit risk early warning method for a market entity, including:

[0007] Determine at least one target early warning indicator that needs to be monitored for the target market entity;

[0008] Based on each of the target early warning indicators, obtain indicator data corresponding to the target early warning indicator, and perform risk early warning monitoring on the target market entity based on the adaptively adjusted early warning threshold and the indicator data;

[0009] Among them, the warning threshold is determined based on the relationship between historical indicator data and historical risk data. The historical indicator data is the indicator data within a preset historical period corresponding to the target warning indicator, and the historical risk data is the risk assessment data of the target market entity within the preset historical period.

[0010] In an embodiment of the present application, by adaptively adjusting the warning threshold based on the relationship between historical indicator data and risk data, the warning threshold can be adjusted in a timely manner to adapt to changes in the relationship between indicators and risks, thereby improving the accuracy of credit risk warnings for market entities.

[0011] In some possible embodiments, the method for adaptively adjusting the warning threshold includes:

[0012] In response to a threshold adjustment instruction for the target early warning indicator, determining early warning indicator variable data corresponding to the target early warning indicator based on the historical indicator data, and determining risk variable data corresponding to the target market entity based on the historical risk data;

[0013] Based on the warning indicator variable data as an independent variable and the risk variable data as a dependent variable, a segmented regression model corresponding to the warning indicator variable data and the risk variable data is constructed, and the warning threshold at the current moment is determined based on the segmentation points of the segmented regression model.

[0014] In an embodiment of the present application, by triggering adaptive adjustment of the threshold in response to a threshold adjustment instruction, and using the segmentation points of the segmented regression model based on indicator variable data and risk variable data as independent variables and dependent variables respectively to determine the current warning threshold, the accuracy of credit risk warning is further improved.

[0015] In some possible embodiments, determining the early warning indicator variable data corresponding to the target early warning indicator based on the historical indicator data includes:

[0016] The historical indicator data is converted into a time series signal, and the time series signal is denoised using a wavelet shrinkage denoising method to obtain early warning indicator variable data.

[0017] In the embodiment of the present application, the historical indicator data is denoised by using a wavelet function to obtain warning indicator variable data for determining the warning threshold, thereby further improving the accuracy of the warning threshold adjustment.

[0018] In some possible embodiments, determining the risk variable data corresponding to the target market entity based on the historical risk data includes:

[0019] Based on the risk characteristic data of multiple dimensions of the target market entity in the preset historical period, the risk characteristic data of the multiple dimensions are weighted and superimposed using a preset weight rule to obtain the risk variable data of the target market entity in the preset historical period.

[0020] In an embodiment of the present application, risk variable data is obtained by weighted superposition of risk characteristic data based on multiple dimensions, thereby further improving the accuracy of early warning threshold adjustment.

[0021] In some possible embodiments, converting the historical indicator data into a time series signal, and performing denoising processing on the time series signal using a wavelet shrinkage denoising method to obtain early warning indicator variable data includes:

[0022] Convert the historical indicator data into a time series signal, perform wavelet decomposition on the time series signal based on a preset wavelet basis function, and obtain a plurality of wavelet coefficients corresponding to the wavelet basis function;

[0023] Performing threshold processing on the plurality of wavelet coefficients to obtain at least one processed target wavelet coefficient;

[0024] Wavelet reconstruction is performed based on the at least one target wavelet coefficient to obtain early warning indicator variable data.

[0025] In the embodiment of the present application, the historical indicator data is denoised by wavelet decomposition and threshold-based processing, thereby further improving the accuracy of the warning threshold adjustment.

[0026] In some possible embodiments, the step of constructing a segmented regression model corresponding to the early warning indicator variable data and the risk variable data based on the early warning indicator variable data as an independent variable and the risk variable data as a dependent variable, and determining the early warning threshold at the current moment based on the segmentation points of the segmented regression model includes:

[0027] Based on the early warning indicator variable data as an independent variable and based on the risk variable data as a dependent variable, construct a segmented regression model corresponding to the early warning indicator variable data and the risk variable data;

[0028] Decomposing the piecewise regression model into a plurality of unit regression models including two consecutive piecewise functions, and selecting a target unit regression model with the highest goodness of fit from the plurality of unit regression models;

[0029] The warning threshold at the current moment is determined based on the segmentation points of the target unit regression model.

[0030] In an embodiment of the present application, a piecewise regression model for indicator variable data and risk variable data is constructed, and two consecutive piecewise functions with the highest goodness of fit are screened to determine the warning threshold based on their segmentation points, thereby further improving the accuracy of the warning threshold adjustment.

[0031] In some possible embodiments, determining at least one target early warning indicator that needs to be monitored for the target market entity includes:

[0032] Based on the multiple candidate indicators of the target market entity, respectively obtain the candidate indicator data corresponding to each candidate indicator within a preset evaluation period, and obtain the credit risk data of the target market entity within the preset evaluation period;

[0033] The correlation coefficients between the candidate indicator data of each candidate indicator and the credit risk data are obtained respectively, and at least one candidate indicator whose correlation coefficient meets a preset condition is determined as at least one target early warning indicator that needs to be monitored currently.

[0034] In the embodiment of the present application, the target warning indicators that need to be warned are screened according to the correlation between the indicator data of multiple candidate indicators and the credit risk data, thereby further improving the accuracy of credit risk warning for market entities.

[0035] In a second aspect, the embodiment of the present application provides a credit risk early warning device for a market entity, including:

[0036] An indicator determination module is used to determine at least one target early warning indicator that needs to be monitored for the target market entity;

[0037] A risk warning module, for obtaining indicator data corresponding to the target warning indicator based on each of the target warning indicators, and performing risk warning monitoring on the target market entity based on the adaptively adjusted warning threshold and the indicator data;

[0038] Among them, the warning threshold is determined based on the relationship between historical indicator data and historical risk data. The historical indicator data is the indicator data within a preset historical period corresponding to the target warning indicator, and the historical risk data is the risk assessment data of the target market entity within the preset historical period.

[0039] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the method described in any embodiment of the first aspect when executing the program.

[0040] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described in any embodiment of the first aspect can be implemented.

[0041] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a computer program, wherein when the computer program is executed by a processor, the method described in any embodiment of the first aspect can be implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A flowchart of a credit risk early warning method for a market entity provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a credit risk early warning device for a market entity provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0047] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0048] It should be noted that the existing technology mainly implements risk warning for monitored objects by setting fixed thresholds for monitoring indicators. The setting of warning thresholds is highly dependent on business experience on the one hand, and is simply adjusted according to the distribution of indicator data on the other hand. For example, the initial threshold is first determined based on business experience, and then the threshold is fine-tuned according to the proportion of objects reaching the threshold to determine the final threshold. It can be seen that although the existing technology has a process of fine-tuning the threshold, the final threshold is still highly dependent on business experience, and the solution of fine-tuning the threshold based solely on the proportion of objects reaching the threshold has the disadvantage of poor adaptability; in addition, because the existing technology adopts a unified warning threshold for all regulatory objects, it does not take into account the individual characteristic differences of the regulatory objects, so the risk warning monitoring lacks pertinence and accuracy.

[0049] In summary, based on the existing credit risk early warning scheme, it is easy to judge the regulatory objects that are actually normal as high-risk, or it is impossible to identify the regulatory objects that are actually high-risk in time, and the accuracy of credit risk early warning is not high.

[0050] In response to the problems existing in the above-mentioned prior art, an embodiment of the present application provides a credit risk warning method for market entities. By performing risk warning monitoring based on an adaptively adjusted warning threshold, the warning threshold can be adjusted in time to adapt to changes in the relationship between indicators and risks, thereby effectively improving the accuracy of credit risk warnings for market entities.

[0051] like Figure 1 As shown, the embodiment of the present application provides a credit risk early warning method for a market entity, which may include the steps of:

[0052] S1. Determine at least one target early warning indicator that needs to be monitored for the target market entity.

[0053] Specifically, step S1 is mainly to determine the types of indicators that currently need to be monitored for risk warning. As the market environment changes or regulatory requirements change, the types of indicators that need to be monitored for the same market entity must also be adjusted accordingly. Therefore, this step can be triggered based on specific conditions. For example, the step of determining the target warning indicator is performed regularly according to a set cycle, or this step can be triggered based on specific events occurring in the market economy environment.

[0054] In step S1, one or more indicators may be selected from multiple candidate indicators according to the adjustment of the regulatory measures or the characteristics of the market entities as at least one target early warning indicator that needs to be monitored currently. It should be noted that the target market entity mainly refers to natural persons, legal persons and unincorporated organizations engaged in business activities for the purpose of profit.

[0055] S2. Based on each target early warning indicator, obtain the indicator data corresponding to the target early warning indicator, and conduct risk early warning monitoring on the target market entity based on the adaptively adjusted early warning threshold and indicator data.

[0056] Among them, the warning threshold is determined based on the relationship between historical indicator data and historical risk data. The historical indicator data is the indicator data within the preset historical period corresponding to the target warning indicator, and the historical risk data is the risk assessment data of the target market entity within the preset historical period.

[0057] Specifically, for each target warning indicator that currently needs to be monitored, the indicator data of this indicator type is obtained in real time, so as to perform risk warning monitoring on the target market entities based on the adaptively adjusted warning threshold and indicator data. If the indicator data of at least one target warning indicator exceeds the corresponding warning threshold, a risk warning is triggered.

[0058] It should be noted that in the process of risk early warning monitoring of the target market entity, the early warning threshold is adaptively adjusted, for example, the adjustment of the early warning threshold can be triggered based on a preset adjustment cycle. In each adjustment process, the historical indicator data corresponding to the target early warning indicator in the historical period and the historical risk data corresponding to the target market entity can be obtained based on the preset historical period ending at the current moment, wherein the historical risk data can be risk assessment data describing the credit risk of the market entity based on the preset feature dimension, for example, the market entity is scored for risk based on administrative penalties, problems found during inspections, abnormal operations and other characteristics. Then, according to the relationship between the historical indicator data and the historical risk data, the early warning threshold is adaptively adjusted. For example, the first mean of the historical indicator data and the second mean of the historical risk data can be calculated respectively, and then the early warning threshold is adjusted according to the relationship between the ratio of the first mean to the second mean and the preset ratio. For example, if it is greater than the preset ratio, the early warning threshold is increased, and if it is less than the preset ratio, the early warning threshold is reduced.

[0059] Based on this, the warning threshold is adaptively adjusted based on the relationship between historical indicator data and risk data, so that the adjusted warning threshold can adapt to the changes in the relationship between indicators and risks, avoiding the misjudgment and omission problems caused by risk monitoring according to the warning threshold fixed based on business experience, thereby effectively improving the accuracy of credit risk warnings for market entities.

[0060] It should be noted that the target market entity of the embodiment of the present application may refer to a single market entity, or a collection of market entities of a certain type or within a certain range. Among them, for the risk warning of a single market entity, the warning threshold can be determined based on the historical indicators and historical risk data of the market entity; the present application scheme can also be applied to the risk warning of a group (a collection of market entities), and the warning threshold can be calculated based on the indicators and risk data of the group accordingly. In addition, when applied to risk warning for a group, the threshold can also be adjusted for each individual in the group. For example, for a certain indicator, the initial threshold of the individual and the initial threshold of the group can be calculated first, and then the individual threshold and the group threshold are weighted and combined to obtain the final warning threshold.

[0061] In some possible embodiments, the early warning indicators can also be classified first, and appropriate threshold setting methods can be used according to different indicator types to adaptively adjust the early warning thresholds of the corresponding indicators. Exemplarily, according to the relationship between early warning indicators and risks, the indicators can be divided into three categories: the first category is positive indicators, which means that the larger the indicator value, the lower the risk, such as the length of time since the establishment of the enterprise. The longer the establishment time of the enterprise, the lower the credit risk of the enterprise. The second category is reverse indicators, which means that the smaller the indicator value, the lower the risk, such as the proportion of administrative penalties for related entities. The third category is stable indicators, which means that the indicator value is best at a certain point or a certain interval. The farther away from this point or this interval, the higher the risk, such as the debt-to-asset ratio.

[0062] In some possible embodiments, the method for adaptively adjusting the warning threshold may include:

[0063] S201, in response to a threshold adjustment instruction for a target early warning indicator, determining early warning indicator variable data corresponding to the target early warning indicator based on historical indicator data, and determining risk variable data corresponding to the target market entity based on historical risk data;

[0064] S202: Based on the early warning indicator variable data as an independent variable and the risk variable data as a dependent variable, a segmented regression model corresponding to the early warning indicator variable data and the risk variable data is constructed, and the early warning threshold at the current moment is determined based on the segmentation points of the segmented regression model.

[0065] Specifically, the triggering conditions for adjusting the warning threshold can be set, for example, only when the threshold adjustment instruction for the target warning indicator is responded to, the warning threshold of the target warning indicator is adjusted. Exemplarily, the threshold adjustment instruction can be initiated manually according to actual needs, or it can be automatically triggered. The automatic triggering condition can be triggered according to a preset adjustment cycle, or triggered according to the occurrence of a specific key event monitored. For example, the specific key event can be that the index fluctuation of a stock market is greater than the set value, or that changes in regulatory rules are monitored, etc.

[0066] Then, the corresponding early warning indicator variable data can be determined based on the acquired historical indicator data, and the corresponding risk variable data can be determined based on the historical risk data. Exemplarily, when the preset historical period is the past year, the historical indicator data of the past year is divided according to the unit time (such as month, week or day, etc.) and converted into the form of a time series to obtain the corresponding early warning indicator variable data; similarly, the historical risk data is divided and converted according to the same rules to obtain the corresponding risk variable data. It can be understood that, since the preset historical time is the same and the unit time of the division is the same, the early warning indicator variable data and the risk variable data are two variable sequence data of equal length, and the data points in the two variable sequence data correspond one to one.

[0067] Based on this, a piecewise regression model is constructed in the form of warning indicator variable data as independent variables and risk variable data as dependent variables (for example, warning indicator variable data corresponds to the X-axis, and risk variable data corresponds to the Y-axis), and several piecewise functions of the warning indicator variable data and the risk variable data are obtained. Then, the warning threshold is determined according to the segmentation points between these segmentation functions. For example, a segmentation point located in a preset range is selected from several segmentation points as the warning threshold.

[0068] In some possible embodiments, in step S201, determining the early warning indicator variable data corresponding to the target early warning indicator based on the historical indicator data may include:

[0069] S2011. Convert the historical indicator data into a time series signal, and use the wavelet shrinkage denoising method to denoise the time series signal to obtain the early warning indicator variable data.

[0070] Further, step S2011 may include:

[0071] Convert historical indicator data into time series signals, perform wavelet decomposition on the time series signals based on a preset wavelet basis function, and obtain a number of wavelet coefficients corresponding to the wavelet basis function;

[0072] Performing threshold processing on a plurality of wavelet coefficients to obtain at least one processed target wavelet coefficient;

[0073] Wavelet reconstruction is performed based on at least one target wavelet coefficient to obtain early warning indicator variable data.

[0074] It should be noted that, considering that there may be some noise in the historical indicator data, the historical indicator data can be denoised to improve the accuracy of risk warning monitoring. Specifically, the wavelet shrinkage denoising method can be used to first convert the historical indicator data into a time series form, and then a wavelet basis function with N layers is selected to perform wavelet decomposition on the time series signal to obtain the corresponding multiple wavelet coefficients.

[0075] For the threshold processing step, a critical threshold can be set. If the wavelet coefficient is less than the critical threshold, it is considered that the coefficient is mainly caused by noise, and this part of the coefficient is removed. If the wavelet coefficient is not less than the critical threshold, it is considered that the coefficient is mainly caused by the signal itself, and this part of the coefficient is retained, and finally at least one target wavelet coefficient is retained; wavelet reconstruction (inverse wavelet transform) is performed based on these target wavelet coefficients to obtain a denoised sequence signal, and this sequence signal is used as the early warning indicator variable data.

[0076] It should be noted that, in addition to directly removing the wavelet coefficients smaller than the critical threshold, the wavelet coefficients smaller than the critical threshold may also be attenuated, and then the attenuated wavelet coefficients are also retained as target wavelet coefficients.

[0077] In some possible embodiments, in step S201, determining risk variable data corresponding to the target market entity based on historical risk data includes:

[0078] Based on the risk characteristic data of multiple dimensions of the target market entities within a preset historical period, the risk characteristic data of multiple dimensions are weighted and superimposed using preset weight rules to obtain the risk variable data of the target market entities within the preset historical period.

[0079] It should be noted that historical risk data can include risk characteristic data of multiple dimensions, such as risk characteristic data of three dimensions, including administrative penalties, problems found during inspections, and abnormal operations. Based on the same preset historical time and the same unit time division rules mentioned above, the risk characteristic data of each dimension can be converted into corresponding sequence data, and then these sequence data are weighted and superimposed using the preset weight rules (each sequence data corresponds to a weight value), and finally the risk variable data of the target market entity in the preset historical period is obtained.

[0080] Based on this, by describing the credit risk situation of market entities from the characteristics of multiple dimensions and weighted superposition of these risk characteristic data of different dimensions to obtain risk variable data, it is possible to more accurately capture the relationship between indicator data and risk, thereby further improving the accuracy of early warning threshold adjustment.

[0081] In some possible embodiments, step S202, based on the early warning indicator variable data as an independent variable and based on the risk variable data as a dependent variable, constructing a segmented regression model corresponding to the early warning indicator variable data and the risk variable data, and determining the early warning threshold at the current moment based on the segmentation points of the segmented regression model, may include:

[0082] S2021. Based on the early warning indicator variable data as an independent variable and the risk variable data as a dependent variable, a segmented regression model corresponding to the early warning indicator variable data and the risk variable data is constructed;

[0083] S2022. decomposing the piecewise regression model into a plurality of unit regression models including two consecutive piecewise functions, and selecting a target unit regression model with the highest goodness of fit from the plurality of unit regression models;

[0084] S2023. Determine the warning threshold at the current moment based on the segmentation points of the target unit regression model.

[0085] It should be noted that the rationality of setting the early warning threshold is directly related to whether the risk early warning system can play a key role in early detection and early identification of risks. At the same time, a reasonably set early warning threshold can avoid the problem of misjudgment. By using the segmented regression model, the threshold effect between the early warning indicator and the risk variable can be explored. The threshold is the value that is most likely to judge the risk of illegal and irregular behavior of market entities, and is defined as the critical value of the early warning indicator for the risk of illegal and irregular behavior of market entities. In theory, this threshold should have the highest sensitivity and specificity: on the one hand, it can identify the market entities with potential illegal and irregular risks to the greatest extent, thereby improving the short-term prediction accuracy of the subject risk; on the other hand, it will not mistakenly judge normal entities as high-risk.

[0086] Specifically, firstly, the early warning indicator variable data is used as the independent variable and the risk variable data is used as the dependent variable to construct the corresponding segmented regression model. The segmented regression model is equivalent to a set of several segmented functions, and each segmented function corresponds to a fitting curve segment with a linear relationship. Then, any two consecutive segmented functions are used as unit regression models, and the goodness of fit of each unit regression model (i.e., regression coefficient R 2 ), where the goodness of fit of each unit regression model can be calculated by the sum of the goodness of fit of the corresponding two-segment segmented functions; finally, the target unit regression model with the highest goodness of fit is selected, and the segmentation point of the two segmented functions in the target unit regression model is used as the final warning threshold. It can be understood that there is a significant difference in the impact of the indicator data on the risk before and after the segmentation point.

[0087] For example, assuming that it is necessary to determine the warning threshold of the target warning indicator of "rapid enterprise expansion warning", first obtain the indicator data of this indicator in the historical period (for example, obtain data such as the number of newly added holding companies and branches in the past year to measure the expansion speed of the enterprise), and use this as the independent variable; then obtain the data measuring the enterprise risk such as the credit risk classification score of the enterprise, and use this as the dependent variable; finally, construct a segmented regression model and determine the warning threshold of the indicator based on the segmentation points.

[0088] In some possible embodiments, step S1, determining at least one target early warning indicator that needs to be monitored for the target market entity, may include:

[0089] S101, based on multiple candidate indicators of the target market entity, respectively obtaining candidate indicator data corresponding to each candidate indicator within a preset evaluation period, and obtaining credit risk data of the target market entity within the preset evaluation period;

[0090] S102: respectively obtain the correlation coefficient between the candidate indicator data and the credit risk data of each candidate indicator, and determine at least one candidate indicator whose correlation coefficient meets a preset condition as at least one target early warning indicator that needs to be monitored currently.

[0091] Specifically, we can first refer to relevant laws, regulations, management methods and rules to clarify the key regulatory matters that need attention, and based on this, determine one or more candidate indicators corresponding to the target market entities.

[0092] When there are multiple candidate indicators, the candidate indicator data of each candidate indicator in the past period of time (preset evaluation period) is obtained respectively, and the credit risk data of the target market entity in the same period of time is obtained at the same time; wherein the credit risk data may include market entity administrative penalty data, market entity administrative inspection data, market entity credit risk score data, etc. Similarly, for these credit risk data and candidate indicator data, they can be divided and converted into time series based on the preset unit time; then the correlation coefficient between the two series data is calculated, for example, the Pearson correlation coefficient of the two series data is calculated.

[0093] After obtaining the correlation coefficient between the candidate indicator data and the credit risk data of each candidate indicator, at least one candidate indicator whose correlation coefficient meets the preset conditions is determined as the target warning indicator that needs to be monitored currently. For example, one or more (the number is set according to the needs) candidate indicators with the largest correlation coefficients can be determined as the target warning indicators that need to be monitored currently; or, the candidate indicators whose correlation coefficients are greater than the preset coefficient threshold can be determined as the target warning indicators that need to be monitored currently.

[0094] Based on this, by selecting the target warning indicators that need to be warned according to the strength of the correlation between the indicator data and the credit risk, we can avoid the misjudgment caused by risk warnings based on indicators with weak correlation, thereby further improving the accuracy of credit risk warnings for market entities.

[0095] It should be noted that, compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0096] 1. The existing scheme adopts a fixed warning threshold. The embodiment of the present application analyzes the relationship between indicators and risks through segmented regression, and adaptively adjusts the warning threshold so that the adjusted warning threshold can adapt to the changes in the relationship between indicators and risks, reducing the reliance on subjective experience, thereby improving the accuracy of risk warning for market entities.

[0097] 2. Existing solutions adopt unified warning thresholds and warning rules for all enterprises or a certain large group, which lacks pertinence and accuracy. The solution of the embodiment of the present application can be applied not only to risk warning monitoring for groups, but also to individual market entities, increasing the consideration of individual characteristics, thereby enhancing the pertinence and accuracy of risk warning for individuals.

[0098] Please refer to Figure 2 , Figure 2 The following is a block diagram showing the composition of the credit risk early warning device of the market subject provided by some embodiments of the present application. It should be understood that the credit risk early warning device of the market subject is similar to the above Figure 1 Corresponding to the method embodiment, the various steps involved in the above method embodiment can be executed. The specific functions of the credit risk early warning device of the market entity can be found in the description above. To avoid repetition, the detailed description is appropriately omitted here.

[0099] Figure 2 The credit risk early warning device of the market subject includes at least one software function module that can be stored in a memory in the form of software or firmware or fixed in the credit risk early warning device of the market subject, and the credit risk early warning device of the market subject includes:

[0100] The indicator determination module 210 is used to determine at least one target early warning indicator that needs to be monitored for the target market entity;

[0101] The risk warning module 220 is used to obtain the indicator data corresponding to the target warning indicator based on each target warning indicator, and perform risk warning monitoring on the target market entity based on the adaptively adjusted warning threshold and indicator data;

[0102] Among them, the warning threshold is determined based on the relationship between historical indicator data and historical risk data. The historical indicator data is the indicator data within the preset historical period corresponding to the target warning indicator, and the historical risk data is the risk assessment data of the target market entity within the preset historical period.

[0103] It can be understood that the above-mentioned device item embodiment corresponds to the method item embodiment of the present invention. The credit risk warning device for a market entity provided by the embodiment of the present invention can implement the credit risk warning method for a market entity provided by any method item embodiment of the present invention.

[0104] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method, and will not be described in detail here.

[0105] like Figure 3 As shown, some embodiments of the present application provide an electronic device 300, which includes: a memory 310, a processor 320, and a computer program stored in the memory 310 and executable on the processor 320, wherein the processor 320 reads the program from the memory 310 through the bus 330 and executes the program to implement a method of any embodiment of the credit risk early warning method for market entities as described above.

[0106] Processor 320 can process digital signals and can include various computing structures, such as complex instruction set computer structure, reduced instruction set computer structure, or a structure that implements a combination of multiple instruction sets. In some examples, processor 320 can be a microprocessor.

[0107] The memory 310 may be used to store instructions executed by the processor 320 or data related to the execution of instructions. These instructions and / or data may include codes for implementing some or all functions of one or more modules described in the embodiments of the present application. The processor 320 of the disclosed embodiment may be used to execute instructions in the memory 310 to implement the method shown above. The memory 310 includes a dynamic random access memory, a static random access memory, a flash memory, an optical memory, or other memory known to those skilled in the art.

[0108] Some embodiments of the present application further provide a computer-readable storage medium having a computer program stored thereon. The computer program is executed by a processor to execute the method described in the method embodiment.

[0109] Some embodiments of the present application further provide a computer program product, which, when executed on a computer, enables the computer to execute the method described in the method embodiment.

[0110] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0111] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0112] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0113] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0114] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0115] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0116] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

Claims

1. A credit risk early warning method for a market entity, characterized in that: include: Determine at least one target early warning indicator that needs to be monitored for the target market entity; Based on each of the target early warning indicators, obtain indicator data corresponding to the target early warning indicator, and perform risk early warning monitoring on the target market entity based on the adaptively adjusted early warning threshold and the indicator data; Among them, the warning threshold is determined based on the relationship between historical indicator data and historical risk data. The historical indicator data is the indicator data within a preset historical period corresponding to the target warning indicator, and the historical risk data is the risk assessment data of the target market entity within the preset historical period.

2. The credit risk early warning method for market entities according to claim 1, characterized in that: The method for adaptively adjusting the warning threshold comprises: In response to a threshold adjustment instruction for the target early warning indicator, determining early warning indicator variable data corresponding to the target early warning indicator based on the historical indicator data, and determining risk variable data corresponding to the target market entity based on the historical risk data; Based on the warning indicator variable data as an independent variable and the risk variable data as a dependent variable, a segmented regression model corresponding to the warning indicator variable data and the risk variable data is constructed, and the warning threshold at the current moment is determined based on the segmentation points of the segmented regression model.

3. The credit risk early warning method for market entities according to claim 2, characterized in that: The determining of the early warning indicator variable data corresponding to the target early warning indicator based on the historical indicator data includes: The historical indicator data is converted into a time series signal, and the time series signal is denoised using a wavelet shrinkage denoising method to obtain early warning indicator variable data.

4. The credit risk early warning method for market entities according to claim 2, characterized in that: The determining of the risk variable data corresponding to the target market entity based on the historical risk data includes: Based on the risk characteristic data of multiple dimensions of the target market entity in the preset historical period, the risk characteristic data of the multiple dimensions are weighted and superimposed using a preset weight rule to obtain the risk variable data of the target market entity in the preset historical period.

5. The credit risk early warning method for market entities according to claim 3 is characterized in that: The historical indicator data is converted into a time series signal, and the time series signal is denoised using a wavelet shrinkage denoising method to obtain early warning indicator variable data, including: Convert the historical indicator data into a time series signal, perform wavelet decomposition on the time series signal based on a preset wavelet basis function, and obtain a plurality of wavelet coefficients corresponding to the wavelet basis function; Performing threshold processing on the plurality of wavelet coefficients to obtain at least one processed target wavelet coefficient; Wavelet reconstruction is performed based on the at least one target wavelet coefficient to obtain early warning indicator variable data.

6. The credit risk early warning method for market entities according to claim 2, characterized in that: The step of constructing a segmented regression model corresponding to the early warning indicator variable data and the risk variable data based on the early warning indicator variable data as an independent variable and the risk variable data as a dependent variable, and determining the early warning threshold at the current moment based on the segmentation points of the segmented regression model includes: Based on the early warning indicator variable data as an independent variable and based on the risk variable data as a dependent variable, construct a segmented regression model corresponding to the early warning indicator variable data and the risk variable data; Decomposing the piecewise regression model into a plurality of unit regression models including two consecutive piecewise functions, and selecting a target unit regression model with the highest goodness of fit from the plurality of unit regression models; The warning threshold at the current moment is determined based on the segmentation points of the target unit regression model.

7. The credit risk early warning method for market entities according to any one of claims 1 to 6, characterized in that: The step of determining at least one target early warning indicator that needs to be monitored for the target market entity includes: Based on the multiple candidate indicators of the target market entity, respectively obtain the candidate indicator data corresponding to each candidate indicator within a preset evaluation period, and obtain the credit risk data of the target market entity within the preset evaluation period; The correlation coefficients between the candidate indicator data of each candidate indicator and the credit risk data are obtained respectively, and at least one candidate indicator whose correlation coefficient meets a preset condition is determined as at least one target early warning indicator that needs to be monitored currently.

8. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the credit risk early warning method for market entities described in any one of claims 1-7 when executing the program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the credit risk early warning method for a market entity as described in any one of claims 1 to 7 is executed.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the credit risk early warning method for a market entity described in any one of claims 1 to 7 is implemented.

Images