Enterprise credit risk model management method, equipment, medium and program product

Abstract

The embodiment of the invention provides an enterprise credit risk model management method and device, a medium and a program product, and relates to the technical field of model adjustment. The method comprises the following steps: determining that a target object credit risk model meets a preset migration matrix updating condition, and obtaining corresponding probability adjustment data based on at least one predetermined external influence factor; and updating a migration matrix of the target object credit risk model based on the probability adjustment data. According to the embodiment of the invention, the updating condition is set, and the migration matrix of the credit risk model is updated in time under the condition that the updating condition is met, so that the updated credit risk model can adapt to the dynamic change of the market and the economic environment, and the accuracy of credit risk assessment is effectively improved.

Description

Technical Field

[0001] The present application relates to the field of model adjustment technology, and in particular to a management method, device, medium and program product of an enterprise credit risk model. Background Art

[0002] Corporate credit risk models are quantitative assessment tools used to evaluate corporate default probability and potential losses. These models can help financial institutions, investors and other stakeholders understand the ability of companies to repay debts, thereby making more informed credit decisions and investment strategies.

[0003] The CreditMetrics model is one of the most common credit risk measurement models. As a risk management tool, the CreditMetrics model can be used to calculate the maximum loss that may occur to a credit asset portfolio during the holding period under a certain confidence level. The core of the CreditMetrics model lies in the analysis of credit rating changes. It describes the probability of changes in credit instruments of different credit ratings within a certain period by constructing a rating transfer matrix (migration matrix), thereby predicting changes in the market value of credit instruments.

[0004] However, the traditional CreditMetrics model relies on a static migration matrix and cannot adapt to the dynamic changes in the market and economic environment, resulting in low accuracy in risk assessment. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a management method, device, medium and program product for an enterprise credit risk model to improve the accuracy of risk assessment.

[0006] In a first aspect, an embodiment of the present application provides a method for managing an enterprise credit risk model, including:

[0007] Determine whether the target object credit risk model satisfies the preset migration matrix update conditions, and obtain corresponding probability adjustment data based on at least one predetermined external influencing factor;

[0008] The migration matrix of the target object credit risk model is updated based on the probability adjustment data.

[0009] In an embodiment of the present application, by setting update conditions and timely updating the migration matrix of the credit risk model when the update conditions are met, the updated credit risk model can adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0010] In some possible embodiments, determining whether the target object credit risk model satisfies a preset migration matrix update condition includes:

[0011] Obtain the time difference between the historical update time corresponding to the last update of the migration matrix and the current time in real time;

[0012] When it is determined that the time difference is greater than the preset migration matrix update period, it is determined that the target object credit risk model meets the preset migration matrix update condition.

[0013] In the embodiment of the present application, by updating the migration matrix according to a preset update cycle, the credit risk model can be regularly adjusted to adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0014] In some possible embodiments, determining whether the target object credit risk model satisfies a preset migration matrix update condition includes:

[0015] The event data corresponding to the at least one external influencing factor is monitored in real time, and when it is determined that the event data satisfies a preset event change condition, it is determined that the target object credit risk model satisfies a preset migration matrix update condition.

[0016] In an embodiment of the present application, by continuously monitoring the event data of key external influencing factors and updating the migration matrix when the preset event change conditions are met, the credit risk model can be adjusted in a timely manner to adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0017] In some possible embodiments, the external influencing factors include indicator-type external influencing factors and event-type external influencing factors. The indicator-type external influencing factors include economic growth indicators, monetary policy indicators, fiscal policy indicators, market concentration, industry growth rate, and industry entry and exit barriers. The event-type external influencing factors include legal and regulatory change events, natural disasters and emergencies, and industry technological change events.

[0018] The determining that the event data meets a preset event change condition includes:

[0019] When the event data of the external influencing factors of the indicator class is monitored to exceed the preset indicator threshold, it is determined that the event data meets the preset event change condition;

[0020] When it is monitored that a preset target event occurs in the event data of the event-type external influencing factors, it is determined that the event data meets a preset event change condition.

[0021] In the embodiments of the present application, by monitoring the changes in external influencing factors from multiple aspects such as indicators and events, it is possible to fully and accurately adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0022] In some possible embodiments, the obtaining corresponding probability adjustment data based on at least one predetermined external influencing factor includes:

[0023] Respectively obtain the factor quantization value and factor weight corresponding to each of the external influencing factors; wherein the factor quantization value corresponding to each of the external influencing factors is calculated based on the quantization rule corresponding to each of the external influencing factors, and the factor weight corresponding to each of the external influencing factors is determined based on the correlation analysis result corresponding to each of the external influencing factors;

[0024] A weighted sum is performed based on the factor quantification value and the factor weight to determine probability adjustment data corresponding to the at least one external influencing factor.

[0025] In an embodiment of the present application, corresponding weights are set by distinguishing the relevance of each external influencing factor to the credit rating, and the probability adjustment data of the migration matrix is ​​calculated according to the quantitative value corresponding to each influencing factor and its weight, thereby further improving the accuracy of the adjusted credit risk model.

[0026] In some possible embodiments, after updating the migration matrix of the target object credit risk model based on the probability adjustment data, the method further includes:

[0027] Perform rationality verification on the updated migration matrix based on preset verification rules;

[0028] When the updated migration matrix meets the verification rule, the corresponding model indicators of the target object credit risk model are updated based on the updated migration matrix;

[0029] When the updated migration matrix does not meet the verification rules, at least one of the quantification rules and the factor weights is adjusted, and the probability adjustment data corresponding to the at least one external influencing factor is re-determined based on the adjusted quantification rules and factor weights to re-update the migration matrix of the target object credit risk model.

[0030] In an embodiment of the present application, by performing rationality verification after updating the migration matrix, and readjusting the quantitative rules or factor weights affecting the probability adjustment data calculation when the rationality conditions are not met, until a qualified migration matrix is ​​obtained, the accuracy of the adjusted credit risk model is further improved.

[0031] In some possible embodiments, the method for determining the at least one external influencing factor includes:

[0032] Determine a plurality of candidate external influencing factors corresponding to the target object credit risk model, and collect historical observation data of each candidate external influencing factor in a preset historical period;

[0033] Obtain credit rating change data of the target object's credit risk model during the preset historical period;

[0034] Based on the historical observation data and the credit rating change data, the correlation between each candidate external influencing factor and the target object credit risk model is determined respectively, and the candidate external influencing factor with a correlation higher than a preset threshold is determined as the at least one external influencing factor.

[0035] In an embodiment of the present application, the correlation between multiple candidate external influencing factors and credit ratings is analyzed based on historical observation data, and key external influencing factors with higher correlation are determined to calculate the migration matrix adjustment data, thereby further improving the accuracy of the adjusted credit risk model.

[0036] In a second aspect, an embodiment of the present application provides a management device for an enterprise credit risk model, including:

[0037] A condition monitoring module, used to determine whether the target object credit risk model meets the preset migration matrix update conditions, and obtain corresponding probability adjustment data based on at least one predetermined external influencing factor;

[0038] A matrix updating module is used to update the migration matrix of the target object credit risk model based on the probability adjustment data.

[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 method for managing an enterprise credit risk model provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a management device for an enterprise credit risk model 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 CreditMetrics model is one of the most common corporate credit risk models, and its main components include:

[0049] Risk exposure: refers to the market value of assets.

[0050] Value at Risk (VaR): This is the core output of the model, which represents the maximum loss that may occur to the credit asset portfolio during the holding period under a certain confidence level.

[0051] Credit Rating: A rating that indicates the credit quality of a borrower.

[0052] Grade transfer matrix: Also known as migration matrix, it is a probability matrix that contains the probability of a credit rating transferring from one grade to another within a certain period of time.

[0053] Loss given default (LGD): When a default occurs, the proportion of the loss borne by the creditor to the total credit contract funds.

[0054] Default recovery rate: the proportion of assets that can be recovered after a debtor defaults.

[0055] At present, the CreditMetrics model mainly relies on a static migration matrix. Due to the lack of an adaptive adjustment mechanism for the migration matrix, the model cannot adapt to the dynamic changes in the market and economic environment, resulting in low accuracy in risk assessment.

[0056] To this end, an embodiment of the present application provides a method for managing an enterprise credit risk model. By setting update conditions and timely updating the migration matrix of the credit risk model when the update conditions are met, the updated credit risk model can adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0057] like Figure 1 As shown, the embodiment of the present application provides a method for managing an enterprise credit risk model, which may include the following steps:

[0058] S1. Determine whether the target object credit risk model satisfies a preset migration matrix update condition, and obtain corresponding probability adjustment data based on at least one predetermined external influencing factor.

[0059] It should be noted that the target object is one of multiple objects to be risk assessed, and the objects to be risk assessed refer to all enterprises that need to undergo credit risk assessment. There is a one-to-one correspondence between the credit risk model and the object to be risk assessed. The embodiment of the present application can monitor the credit risk model of each object to be risk assessed in real time. When it is monitored that the credit risk model of the target object meets the preset migration matrix update conditions, it triggers the adjustment and update of the migration matrix of the credit risk model of the target object.

[0060] Specifically, the migration matrix update conditions can be pre-set according to needs, and can be set to be updated periodically, or can be set to trigger updates based on specific events, or can be triggered in response to migration matrix update instructions from administrators.

[0061] When it is determined that the target object credit risk model satisfies the preset migration matrix update condition, corresponding probability adjustment data is obtained based on at least one predetermined external influencing factor corresponding to the target object credit risk model.

[0062] Among them, external influencing factors refer to factors other than the customer's own reasons that may affect its credit risk assessment, such as various influencing factors including the macroeconomic environment, industry competition situation, external events of the enterprise, etc.

[0063] The probability adjustment data refers to a set of values ​​for adjusting the migration matrix, and the probability adjustment data can be calculated based on the change of each external influencing factor and a preset corresponding relationship.

[0064] S2. Update the migration matrix of the target object's credit risk model based on the probability adjustment data.

[0065] After the probability adjustment data corresponding to at least one external influencing factor is obtained, the migration matrix of the credit risk model is adjusted and updated based on the probability adjustment data.

[0066] Based on this, by setting update conditions, the migration matrix of the credit risk model is updated in time when the corresponding conditions are met, so that the credit risk assessment using this model can adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0067] It should be noted that after the migration matrix is ​​updated, other elements or related output results of the target object's credit risk model may be updated and calculated based on the updated migration matrix.

[0068] For example, the default probability of a borrower in the future can be calculated using a dynamically adjusted migration matrix, and the model will be evaluated based on the latest credit rating information. Specifically, first determine the initial position in the migration matrix based on the borrower's current credit rating; then divide the future time period to be evaluated into multiple small time steps; starting from the initial state, for each time step, calculate the probability of being in a different credit rating in the next time step based on the current state and the migration matrix; assuming that in the migration matrix, the lowest rating is defined as the default state, then as the time step progresses, the probability of reaching the default state that accumulates continuously is the default probability of the borrower corresponding to the current credit rating.

[0069] For example, assuming that the initial state is credit rating A, the migration matrix is ​​M, and the probability vectors of different credit ratings after the first time step are P1=P0×M, where P0 is the initial state probability vector (only the position corresponding to credit rating A is 1, and the others are 0). After the second time step, the probability vector is P2=P1×M, and so on. The probability accumulation of the default state is calculated continuously until the default state is reached, and the probability of the final default state is obtained, which is the default probability of the current borrower (target object) in the future time period.

[0070] It should be noted that loss payment refers to the amount of compensation that the borrower needs to pay to the insured or creditor when an agreed loss event occurs in an insurance contract or financial transaction.

[0071] It is understandable that loss payments are closely related to the probability of default and the default loss rate, where the probability of default refers to the possibility that the debtor will not be able to perform the debt contract on time in the future. The higher the probability of default, the greater the possibility of loss events, thereby increasing the possibility and frequency of loss payments. For example, in credit risk management, if a borrower has a high probability of default, then the financial institution may need to prepare for potential default losses, which may include increasing reserves to cope with possible loss payments. A higher probability of default may also cause financial institutions to increase interest rates or fees when pricing loans or other financial products to compensate for potential loss risks.

[0072] In addition, the default loss rate refers to the severity of the loss in the event of a debtor default, usually expressed as the ratio of the loss amount at the time of default to the debt amount. The higher the default loss rate, the greater the amount of loss payment may be in the event of a default. For example, in corporate credit risk assessment, if a company has a high default loss rate, it may be due to its poor asset quality, insufficient collateral, or high industry risk. In this case, the creditor may suffer a greater loss when facing a default, and thus require a higher loss payment to make up for the loss.

[0073] For example, the expected loss (EL) can be calculated based on the default probability, default loss rate and risk exposure corresponding to the current model. Specifically, the following formula can be used to express the relationship between loss payment (expected loss), default probability and default loss rate: Expected loss = default probability × default loss rate × risk exposure. Among them, expected loss refers to the average loss amount that may be incurred due to the debtor's default in the future; risk exposure refers to the amount of loss faced by the creditor when a default event occurs.

[0074] It can be understood that the size of the loss payment depends on three factors: the probability of default, the default loss rate and the risk exposure. The higher the probability of default and the default loss rate, the greater the risk exposure, the higher the expected loss, which may lead to higher loss payments.

[0075] In actual applications, financial institutions and enterprises usually determine the risk level by evaluating the probability of default and the loss rate given default, and take corresponding risk management measures, such as adjusting credit limits, strengthening risk monitoring, purchasing insurance, etc., to reduce the possibility and amount of loss payments. At the same time, expected losses can also be taken into account in cost considerations through reasonable pricing of financial products or services to ensure the sustainability and profitability of the business.

[0076] It should be noted that the credit risk model based on the dynamic migration matrix can also calculate the VaR (Value at Risk) caused by credit risk, which represents the maximum loss that may be incurred at a specific confidence level. Since VaR is calculated based on the dynamically updated migration matrix, the calculation results can reflect market changes and corporate credit status in real time, with high adaptability and timeliness, thereby improving the accuracy of credit risk assessment.

[0077] In some possible embodiments, in step S1, determining whether the target object credit risk model satisfies a preset migration matrix update condition may include:

[0078] S101, obtaining in real time the time difference between the historical update time corresponding to the last update of the migration matrix and the current time;

[0079] S102: When it is determined that the time difference is greater than a preset migration matrix update period, determining whether the target object credit risk model satisfies a preset migration matrix update condition.

[0080] Specifically, the cycle for updating the migration matrix (preset migration matrix update cycle) can be set according to actual needs, such as monthly, weekly, etc.; then, the time difference between the time when the migration matrix was last updated and the current time is obtained in real time based on the time recorder (equivalent to the accumulated time when the migration matrix is ​​not updated). When the time difference is greater than the set migration matrix update cycle, it is determined that the preset migration matrix update condition is met.

[0081] Based on this, by updating the migration matrix according to the preset update cycle, the credit risk model can be regularly adjusted to adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0082] In some possible embodiments, in step S1, determining whether the target object credit risk model satisfies a preset migration matrix update condition may include:

[0083] S111. Monitor event data corresponding to at least one external influencing factor in real time, and when it is determined that the event data satisfies a preset event change condition, determine whether the target object credit risk model satisfies a preset migration matrix update condition.

[0084] It should be noted that, in addition to updating the migration matrix regularly, the migration matrix can also be updated based on changes in specific events (such as major economic changes, corporate financial crises, etc.).

[0085] Specifically, based on at least one external influencing factor that is predetermined to have a considerable influence on the credit risk model of the target object, the event data of these external influencing factors are obtained, and then it is determined based on these event data whether the preset event change conditions are met. If so, it is determined that the credit risk model of the target object meets the preset migration matrix update conditions, and the migration matrix needs to be updated.

[0086] Based on this, by continuously monitoring the event data of key external influencing factors and updating the migration matrix when the preset event change conditions are met, the credit risk model can be adjusted in time when specific events occur to adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0087] In some possible embodiments, external influencing factors include indicator-type external influencing factors and event-type external influencing factors. Indicator-type external influencing factors include economic growth indicators, monetary policy indicators, fiscal policy indicators, market concentration, industry growth rate, and industry entry and exit barriers. Event-type external influencing factors include legal and regulatory change events, natural disasters and emergencies, and industry technological change events.

[0088] In step S111, the step of determining whether the event data meets the preset event change condition may include:

[0089] S1111. When it is detected that the event data of the external influencing factor of the indicator exceeds the preset indicator threshold, it is determined that the event data meets the preset event change condition;

[0090] S1112: When it is detected that a preset target event occurs in the event data of the event-type external influencing factors, it is determined that the event data satisfies a preset event change condition.

[0091] It should be noted that judging whether the event data satisfies the preset event change conditions can be judging whether the event data (indicator value) corresponding to a certain external influencing factor exceeds a preset threshold (for example, the gross domestic product growth rate exceeds a preset threshold), or judging whether a preset event occurs in the event data corresponding to a certain external influencing factor (for example, a natural disaster event is monitored).

[0092] Specifically, information such as market news, economic indicators and industry dynamics can be integrated to identify external factors that affect credit ratings.

[0093] Exemplarily, indicator-type external influencing factors mainly include relevant indicators in terms of macroeconomic environment, industry competition situation, etc. Among them, the macroeconomic environment can include economic growth indicators, monetary policy indicators, and fiscal policy indicators. The industry competition situation can include market concentration, industry growth rate, and industry entry and exit barriers.

[0094] Exemplarily, economic growth indicators may include the GDP growth rate, industrial added value, and the consumer price index (CPI). Among them, for the GDP growth rate, it is understandable that the overall economic growth has an important impact on the business and credit status of enterprises: in periods of rapid economic growth, enterprises are generally more likely to obtain market opportunities and financing, and their credit ratings may rise; while in periods of economic recession, enterprises face increased market pressure and increased credit risk. For industrial added value, it reflects the production activities of industrial enterprises. For enterprises that rely on industrial production, changes in industrial added value will directly affect their credit ratings. For the consumer price index, it is used to measure the level of inflation. High inflation may lead to rising costs and falling profits for enterprises, thereby affecting credit ratings.

[0095] For example, monetary policy indicators may include interest rate levels and money supply. For interest rate levels, changes in interest rates will affect the financing costs of enterprises: rising interest rates will increase the debt burden of enterprises, reduce their profitability, and may lead to a decline in credit ratings; falling interest rates will help enterprises reduce financing costs and improve credit ratings. For money supply, an increase or decrease in money supply will affect market liquidity and the financing environment of enterprises: when the money supply is sufficient, enterprises are more likely to obtain financing and their credit ratings may rise; when the money supply is tight, it will be more difficult for enterprises to obtain financing and their credit risks will rise.

[0096] For example, fiscal policy indicators can include expenditure and tax policies. Fiscal expenditure policies have different impacts on enterprises in different industries: for example, increasing investment in infrastructure construction may benefit related enterprises and improve their credit ratings; while reducing subsidies for certain industries may make these enterprises face difficulties in operation and reduce their credit ratings. As for tax policies, adjustments to tax policies will affect the profit level of enterprises: reducing the tax burden such as corporate income tax can improve the profitability of enterprises and help improve their credit ratings; increasing taxes may lead to a decline in corporate profits and an increase in credit risk.

[0097] For example, market concentration reflects the number and size distribution of enterprises in the industry. If the industry concentration is high and a few large enterprises dominate, then these enterprises usually have strong market competitiveness and risk resistance, and their credit ratings are relatively high. On the contrary, if the industry competition is fierce, there are many enterprises with small scale and dispersed market share, the credit risk of enterprises may be high.

[0098] For example, industry growth rates reflect changes in market demand. Companies in fast-growing industries usually have more development opportunities and may have higher credit ratings; while companies in slow-growing or declining industries face greater market pressure and higher credit risks.

[0099] For example, the degree of industry entry and exit barriers refers to the difficulty of entering or exiting an industry. For industries with high entry barriers, it is difficult for new enterprises to enter, the competitive pressure on existing enterprises is relatively small, and the credit rating may be high; entry barriers may include technical barriers, capital barriers, policy barriers, etc. For industries with high exit barriers, it is difficult for enterprises to exit the market when they are in financial difficulties, which may lead to an excessive number of enterprises in the industry, intensified competition, and increased credit risk; exit barriers may include asset specificity, labor placement costs, legal and policy restrictions, etc.

[0100] It is understandable that event-type external influencing factors mainly include legal and regulatory change events, natural disasters and emergencies, and industry technology change events. Among them, for legal and regulatory change events: the introduction of new laws and regulations may have a significant impact on the business activities of certain companies. For example, the strengthening of environmental protection regulations may cause high-polluting companies to face huge environmental protection investment and penalty risks, and their credit ratings will decline; adjustments to financial regulatory policies may affect the financing channels and costs of companies, and thus affect their credit ratings. For natural disasters and emergencies, natural disasters such as earthquakes, floods, typhoons, etc., as well as emergencies such as terrorist attacks and public health events, may cause serious damage to the production and operation of certain companies, resulting in a decline in credit ratings. For industry technology change events, rapid technological changes may put some companies at risk of being eliminated, while others may gain competitive advantages through technological innovation. For example, with the development of new energy technologies, the credit risk of traditional energy companies may rise, while the credit rating of new energy companies may improve.

[0101] Based on this, by monitoring the changes in external influencing factors from multiple aspects such as indicators and events, we can fully and accurately adapt to the dynamic changes in the market and economic environment, thereby effectively improving the accuracy of credit risk assessment.

[0102] In some possible embodiments, in step S1, acquiring corresponding probability adjustment data based on at least one predetermined external influencing factor may include:

[0103] S121, respectively obtaining the factor quantization value and factor weight corresponding to each external influencing factor; wherein the factor quantization value corresponding to each external influencing factor is calculated based on the quantization rule corresponding to each external influencing factor, and the factor weight corresponding to each external influencing factor is determined based on the correlation analysis result corresponding to each external influencing factor;

[0104] S122: Perform weighted summation based on the factor quantification value and the factor weight to determine probability adjustment data corresponding to at least one external influencing factor.

[0105] Specifically, first collect the corresponding data of various external influencing factors in the past period of time, such as macroeconomic data, industry data, and enterprise external event data, etc. These data can be obtained from departments, industry associations, financial institutions, news media and other channels.

[0106] Then, for each external influencing factor, the corresponding quantitative value is obtained based on the above-obtained data according to a predetermined quantification method. For example, for the macroeconomic growth rate, it can be divided into different intervals and assigned corresponding quantitative values; for the intensity of industry competition, market concentration indicators can be used for quantification. In addition, for qualitative external influencing factors (such as event-type external influencing factors), such as changes in laws and regulations, natural disasters, etc., they can be quantified through expert evaluation or a pre-established scoring system.

[0107] After obtaining the factor quantization value corresponding to each external influencing factor, these factor quantization values ​​are weighted and summed based on the factor weight corresponding to each external influencing factor to obtain the final probability adjustment data for probability updating of the migration matrix.

[0108] It should be noted that since different external influencing factors have different degrees of relevance to credit ratings, the numerical changes corresponding to each external influencing factor have different degrees of influence on the update of the migration matrix. Therefore, the corresponding factor weights can be set based on the correlation analysis results corresponding to each external influencing factor.

[0109] Specifically, relevant data of a past historical period corresponding to the current moment can be obtained, including event data (indicator data) of each external influencing factor and the corresponding corporate credit rating results. Then, relevant statistical methods (such as correlation analysis, regression analysis, etc.) can be used to determine the direction (positive or negative) and degree of influence of each external influencing factor on the credit rating. According to the degree of influence, a corresponding factor weight can be set for each external influencing factor (the factor weight can also be added with a corresponding positive or negative sign based on the direction of influence of the external influencing factor).

[0110] It is understandable that if external factors have a positive impact on credit ratings, that is, improve the credit status of the enterprise, then the corresponding probability of upward migration should increase, and the probability of downward migration should decrease; conversely, if external factors have a negative impact on credit ratings, then the probability of downward migration should increase, and the probability of upward migration should decrease.

[0111] According to the quantitative value and weight of each external influencing factor, the new migration probability of each credit rating status can be calculated. For example: new migration probability = basic migration probability + external factor impact adjustment value; external factor impact adjustment value = ∑ (factor quantification value i × factor weight i), i is a positive integer less than or equal to n, where n is the number of external influencing factors of the target object credit risk model.

[0112] In some possible embodiments, after step S2, the following steps may also be included:

[0113] S301, verifying the rationality of the updated migration matrix based on preset verification rules;

[0114] S302: When the updated migration matrix meets the verification rules, the corresponding model indicators of the target object credit risk model are updated based on the updated migration matrix;

[0115] S303. When the updated migration matrix does not meet the verification rules, at least one of the quantitative rules and factor weights is adjusted, and the probability adjustment data corresponding to at least one external influencing factor is re-determined based on the adjusted quantitative rules and factor weights to re-update the migration matrix of the target object's credit risk model.

[0116] It should be noted that in order to ensure that the model after the migration matrix is ​​updated conforms to the basic logic, the rationality of the updated migration matrix can be verified. The migration matrix is ​​officially adopted only after the verification is passed. Otherwise, the updated data of the migration matrix is ​​re-obtained after adjusting the relevant values.

[0117] Specifically, the rationality of the updated migration matrix is ​​verified based on the preset verification rules, which can be to determine whether the probability of the updated migration matrix is ​​within a preset probability value range (for example, between 0 and 1), or to determine whether there are unreasonable jumps or mutations in the migration probability, for example, the difference before and after the update is greater than a set threshold.

[0118] When the updated migration matrix does not meet the verification rules, at least one of the quantitative rules and factor weights needs to be adjusted. For example, the quantitative rules and / or factor weights can be adjusted accordingly according to the corresponding adjustment strategy when the preset probability is too large or too small; the corresponding data can also be sent to the administrator terminal so that experts can discuss and refer to similar historical situations to adjust the quantitative rules and / or factor weights and provide feedback.

[0119] Then, based on the adjusted quantitative rules and / or factor weights, the probability adjustment data corresponding to at least one external influencing factor is reacquired to re-update the migration matrix of the target object's credit risk model. The rationality verification and adjustment process is repeated until a migration matrix that meets the preset rationality rules is obtained.

[0120] In some possible embodiments, the method for determining at least one external influencing factor includes:

[0121] Determine multiple candidate external influencing factors corresponding to the target object's credit risk model, and collect historical observation data of each candidate external influencing factor in a preset historical period;

[0122] Obtain credit rating change data of the target object's credit risk model in a preset historical period;

[0123] Based on the historical observation data and the credit rating change data, the correlation between each candidate external influencing factor and the credit risk model of the target object is determined respectively, and the candidate external influencing factor with a correlation higher than a preset threshold is determined as at least one external influencing factor.

[0124] It should be noted that in the process of constructing the target object credit risk model, it is necessary to first determine which external factors have a certain impact on the credit rating of the enterprise.

[0125] Specifically, first, all predetermined influencing factors can be used as multiple candidate external influencing factors to be screened. Then, based on the current moment, historical observation data of these candidate external influencing factors in the past preset historical period are collected respectively, and at the same time, the credit rating change data of the target object credit risk model in the same historical period is obtained.

[0126] Then, based on the preset correlation analysis method, the correlation between each candidate external influencing factor and the target object credit risk model can be obtained respectively, and finally several candidate external influencing factors with the highest correlation (or above the threshold) are determined as at least one external influencing factor corresponding to the target object credit risk model.

[0127] For example, statistical software can be used to analyze the correlation between external factors and corporate credit ratings, such as calculating the Pearson correlation coefficient, Spearman rank correlation coefficient, etc., to find external factors with a high correlation with credit ratings.

[0128] For example, external influencing factors can be used as independent variables and corporate credit ratings as dependent variables, based on which a regression model is constructed to perform regression analysis. The degree (correlation) and direction of the influence of external influencing factors on credit ratings can be determined through regression analysis.

[0129] It should be noted that since the external influencing factors are changing dynamically, it is possible to continuously monitor the changes in the external influencing factors. By establishing a monitoring mechanism, the data changes of the external influencing factors can be regularly collected and analyzed. When a large change is monitored based on the data changes, it triggers the re-determination of at least one external influencing factor of the target object's credit risk model.

[0130] Please refer to Figure 2 , Figure 2 FIG. 1 is a block diagram showing the composition of the management device of the enterprise credit risk model provided by some embodiments of the present application. It should be understood that the management device of the enterprise credit risk model is similar to the above-mentioned Figure 1 Corresponding to the method embodiment, the various steps involved in the above method embodiment can be executed. The specific functions of the management device of the enterprise credit risk model can be found in the description above. To avoid repetition, the detailed description is appropriately omitted here.

[0131] Figure 2 The management device of the enterprise credit risk model 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 management device of the enterprise credit risk model, and the management device of the enterprise credit risk model includes:

[0132] The condition monitoring module 210 is used to determine whether the target object credit risk model meets the preset migration matrix update condition, and obtain corresponding probability adjustment data based on at least one predetermined external influencing factor;

[0133] The matrix updating module 220 is used to update the migration matrix of the target object credit risk model based on the probability adjustment data.

[0134] It can be understood that the above-mentioned device item embodiment corresponds to the method item embodiment of the present invention. The management device of an enterprise credit risk model provided by the embodiment of the present invention can implement the management method of an enterprise credit risk model provided by any method item embodiment of the present invention.

[0135] 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.

[0136] 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 the method of any embodiment of the management method of the enterprise credit risk model as described above.

[0137] 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.

[0138] 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.

[0139] 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.

[0140] 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.

[0141] 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.

[0142] 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.

[0143] 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.

[0144] 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.

[0145] 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.

[0146] 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.

[0147] 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 management method for an enterprise credit risk model, characterized in that: include: Determine whether the target object credit risk model satisfies the preset migration matrix update conditions, and obtain corresponding probability adjustment data based on at least one predetermined external influencing factor; The migration matrix of the target object credit risk model is updated based on the probability adjustment data.

2. The management method of the enterprise credit risk model according to claim 1, characterized in that: The step of determining whether the target object credit risk model satisfies a preset migration matrix update condition includes: Obtain the time difference between the historical update time corresponding to the last update of the migration matrix and the current time in real time; When it is determined that the time difference is greater than the preset migration matrix update period, it is determined that the target object credit risk model meets the preset migration matrix update condition.

3. The management method of enterprise credit risk model according to claim 1, characterized in that: The step of determining whether the target object credit risk model satisfies a preset migration matrix update condition includes: The event data corresponding to the at least one external influencing factor is monitored in real time, and when it is determined that the event data satisfies a preset event change condition, it is determined that the target object credit risk model satisfies a preset migration matrix update condition.

4. The method for managing an enterprise credit risk model according to claim 3, characterized in that: The external influencing factors include indicator-type external influencing factors and event-type external influencing factors. The indicator-type external influencing factors include economic growth indicators, monetary policy indicators, fiscal policy indicators, market concentration, industry growth rate and industry entry and exit barriers. The event-type external influencing factors include legal and regulatory change events, natural disasters and emergencies, and industry technological change events; The determining that the event data meets a preset event change condition includes: When the event data of the external influencing factors of the indicator class is monitored to exceed the preset indicator threshold, it is determined that the event data meets the preset event change condition; When it is monitored that a preset target event occurs in the event data of the event-type external influencing factors, it is determined that the event data meets a preset event change condition.

5. The management method of enterprise credit risk model according to claim 1, characterized in that: The obtaining corresponding probability adjustment data based on at least one predetermined external influencing factor includes: Respectively obtain the factor quantization value and factor weight corresponding to each of the external influencing factors; wherein the factor quantization value corresponding to each of the external influencing factors is calculated based on the quantization rule corresponding to each of the external influencing factors, and the factor weight corresponding to each of the external influencing factors is determined based on the correlation analysis result corresponding to each of the external influencing factors; A weighted sum is performed based on the factor quantification value and the factor weight to determine probability adjustment data corresponding to the at least one external influencing factor.

6. The management method of enterprise credit risk model according to claim 5, characterized in that: After the migration matrix of the target object credit risk model is updated based on the probability adjustment data, the method further includes: Perform rationality verification on the updated migration matrix based on preset verification rules; When the updated migration matrix meets the verification rule, the corresponding model indicators of the target object credit risk model are updated based on the updated migration matrix; When the updated migration matrix does not meet the verification rules, at least one of the quantification rules and the factor weights is adjusted, and the probability adjustment data corresponding to the at least one external influencing factor is re-determined based on the adjusted quantification rules and factor weights to re-update the migration matrix of the target object credit risk model.

7. The method for managing an enterprise credit risk model according to claim 1, characterized in that: The method for determining the at least one external influencing factor includes: Determine a plurality of candidate external influencing factors corresponding to the target object credit risk model, and collect historical observation data of each candidate external influencing factor in a preset historical period; Obtain credit rating change data of the target object's credit risk model during the preset historical period; Based on the historical observation data and the credit rating change data, the correlation between each candidate external influencing factor and the target object credit risk model is determined respectively, and the candidate external influencing factor with a correlation higher than a preset threshold is determined as the at least one external influencing factor.

8. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor can implement the enterprise credit risk model management method described in any one of claims 1 to 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 method for managing an enterprise credit risk model according to 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 method for managing the enterprise credit risk model according to any one of claims 1 to 7 is implemented.

Images