Credit default transfer matrix calibration method and device, equipment and storage medium

Abstract

The invention relates to the technical field of enterprise credit evaluation, in particular to a credit default transfer matrix calibration method and device, equipment and a storage medium, and the method comprises the steps: constructing an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, the credit rating information comprises an initial credit rating, a period rating situation and an initial rating data volume; performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average transfer matrix; calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix; calibrating the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default migration matrix; according to the method and the device, the accuracy of enterprise bond credit rating estimated through the credit default transfer matrix can be conveniently improved.

Description

Technical Field

[0001] The present application relates to the technical field of enterprise credit assessment, and in particular to a credit default transfer matrix calibration method, device, equipment and storage medium. Background Art

[0002] With the rapid expansion of the economy, more and more companies have begun to rely on bond issuance to meet their growing financing needs. Compared with traditional bank loans, bond financing not only provides companies with a more flexible funding channel, but also provides investors with the opportunity to choose between risk and return. Bonds issued by companies are usually classified according to their credit ratings, ranging from lower-risk investment-grade bonds to higher-risk non-investment-grade or high-yield bonds. The higher the credit rating, the lower the risk of default of the company and the greater the confidence of investors in it.

[0003] At present, most traditional credit default transfer matrices are based on historical experience data and Markov chain models, which assume that the change process of corporate credit ratings has a certain stability. However, in a specific market, the change of credit ratings is affected by many factors, including industry characteristics, policy changes, economic fluctuations, etc. These factors work together to make the rating transfer process more complex and changeable. Especially in the case of economic downturn or declining industry prosperity, the default risk of enterprises will rise sharply, and the traditional rating transfer matrix is ​​difficult to capture this nonlinear change. In addition, the channels and data volume for obtaining credit rating data in a specific market may be relatively limited, and market participants are highly dependent on credit ratings. Therefore, in actual applications, the corporate bond credit ratings estimated by the credit default transfer matrix often have inaccurate problems. This not only increases the uncertainty of financial institutions in credit risk management, but also leads to risk judgment bias in investors' bond investments. Therefore, it is necessary to calibrate the credit default transfer matrix according to the actual situation of a specific market. Summary of the invention

[0004] In order to improve the accuracy of corporate bond credit ratings estimated by a credit default transfer matrix, the present application provides a credit default transfer matrix calibration method, apparatus, device and storage medium.

[0005] In a first aspect, the present application provides a credit default transfer matrix calibration method, comprising:

[0006] Constructing an initial probability transfer matrix set based on credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume;

[0007] Performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix;

[0008] Calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix;

[0009] The multi-year average migration matrix is ​​calibrated based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0010] In a second aspect, the present application provides a credit default transfer matrix calibration device, comprising:

[0011] A matrix set construction module, used to construct an initial probability transfer matrix set based on credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume;

[0012] A weighted calculation module, used for performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix;

[0013] A calibration matrix calculation module, used for calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix;

[0014] A matrix calibration module is used to calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0015] In a third aspect, the present application provides a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above method when executing the computer program.

[0016] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps in the above method when executed by a processor.

[0017] In a fifth aspect, the present application further provides a computer program product, wherein the computer program product comprises a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0018] The above-mentioned credit default transfer matrix calibration method, device, equipment and storage medium construct an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume; weighted calculation is performed on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix; an implicit default probability matrix is ​​calculated based on the multi-year average migration matrix, and a calibration parameter matrix is ​​calculated based on the implicit default probability matrix and a preset calibration parameter inverse matrix; the multi-year average migration matrix is ​​calibrated based on the calibration parameter matrix to obtain a target credit default transfer matrix; through the above-mentioned implementation, by introducing multi-dimensional data sources (such as initial credit rating, period rating situation and initial rating data volume, etc.), the impact of different factors on the credit of the enterprise is comprehensively considered, so as to truly reflect the fluctuation of the credit rating of the enterprise; by selecting initial probability transfer matrices of different years for weighted calculation, the target credit default transfer matrix finally calibrated can be not only suitable for the rating change prediction of short-term bonds, but also can provide effective risk assessment for long-term bond investment; furthermore, the use of the target credit default transfer matrix can more accurately predict the credit rating changes of corporate bonds.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention 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 creative work.

[0021] Figure 1 A flow chart of a credit default transfer matrix calibration method provided in an embodiment of the present application;

[0022] Figure 2 A flow chart of another credit default transfer matrix calibration method provided in an embodiment of the present application;

[0023] Figure 3 A schematic diagram of the structure of a credit default transfer matrix calibration device provided in an embodiment of the present application;

[0024] Figure 4 This is a schematic diagram of the structure of another credit default transfer matrix calibration device provided in an embodiment of the present application;

[0025] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present application;

[0026] Figure 6 This is a diagram of the internal structure of a computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present disclosure more clear, the present disclosure is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present disclosure and are not used to limit the present disclosure.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of this article and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this article described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0029] In this article, the term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship.

[0030] Embodiment 1

[0031] Figure 1 A flow chart of a credit default transfer matrix calibration method provided in Example 1 of the present application, referring to Figure 1 The method may be performed by a device for performing the method, and the device may be implemented by software and / or hardware. The method includes:

[0032] S110. Construct an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, interim rating situation, and initial rating data volume.

[0033] Among them, since there are bonds of different maturities on the market, in order to accurately estimate the credit rating of corporate bonds, this embodiment also needs to collect credit rating information of enterprises with multiple maturities; in this embodiment, the multiple maturities preferably include five years and ten years, and in other embodiments, the multiple maturities may also include at least two other different maturities, which are not specifically limited; the initial credit rating is the credit rating of the corporate bond when the credit rating of the corporate bond is monitored within a monitoring period, and the credit rating may include: AAA, AA, A, BBB, BB... in descending order; the period rating situation is the change in the credit rating of the corporate bond within the monitoring period situation; the amount of initial rating data is the number of enterprises selected during the monitoring period; in the example row, the monitoring period is from January 1 to December 31 of a year; the initial probability transfer matrix set includes the initial probability transfer matrix corresponding to each year, and the initial probability transfer matrix is ​​used to estimate the credit rating of corporate bonds. However, due to the poor estimation accuracy of the initial probability transfer matrix, the initial probability transfer matrix needs to be optimized later; the initial probability transfer matrix is ​​constructed based on the credit rating information of enterprises of the corresponding years; each element in the initial probability transfer matrix is ​​used to represent the probability that the corresponding level of credit rating remains unchanged or changes to another level of credit rating during the monitoring period.

[0034] Specifically, credit rating information of enterprises of multiple years is selected, and a one-to-one corresponding initial probability transfer matrix is ​​constructed according to the credit rating information of the enterprise of each year, and the initial probability transfer matrices are assembled to obtain an initial probability transfer matrix set.

[0035] S120, performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix.

[0036] Among them, the multi-year average migration matrix is ​​obtained by weighted calculation of each initial probability transfer matrix in the initial probability transfer matrix set. Since the multi-year average migration matrix integrates the initial probability transfer matrices corresponding to different years, it can show the rating stability of high-grade entities and bonds with relatively rich samples.

[0037] Specifically, weighted calculation is performed on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix.

[0038] S130. Calculate an implicit default probability matrix based on the multi-year average migration matrix, and calculate a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix.

[0039] Among them, each element of the implicit default probability matrix is ​​used to characterize the probability of default of the corresponding credit rating during the monitoring period. The implicit default probability matrix is ​​calculated based on the multi-year average migration matrix. Assuming that the multi-year average migration matrix involving a five-year period is to be calibrated later, the first calibration parameter matrix Q(θ0) corresponding to the 0th year to the 1st year can be calculated based on the implicit default probability matrix, and a first calibration parameter inverse matrix Q(θ0) needs to be preset for the 1st year to the 2nd year. -1 , so as to facilitate the subsequent calculation of the second calibration parameter matrix Q(θ1) corresponding to the first year to the second year; it is also necessary to preset a second calibration parameter inverse matrix Q(θ1) for the second year to the third year. -1 , so as to facilitate the subsequent calculation of the third calibration parameter matrix Q(θ2) corresponding to the 2nd year to the 3rd year; it is also necessary to preset a third calibration parameter inverse matrix Q(θ2) for the 3rd year to the 4th year. -1 , so as to facilitate the subsequent calculation of the fourth calibration parameter matrix Q(θ3) corresponding to the 3rd year to the 4th year; it is also necessary to preset a fourth calibration parameter inverse matrix Q(θ3) for the 4th year to the 5th year. -1 , so as to facilitate the subsequent calculation of the fifth calibration parameter matrix Q(θ4) corresponding to the 3rd year to the 4th year.

[0040] Among them, each element in the first calibration parameter matrix Q(θ0) The calculation formula is: Wherein, I is the identity matrix, DP is the implicit default probability matrix, and P is the multi-year average migration matrix; illustratively, Q(θ0)*Q(θ1) is the first preset value, and further, the first preset value and the first calibration parameter inverse matrix Q(θ0) are calculated: -1 The product of can calculate the second calibration parameter matrix Q(θ1); further, Q(θ1)*Q(θ2) is the second preset value, calculate the second preset value and the second calibration parameter inverse matrix Q(θ1) -1 The third calibration parameter matrix Q(θ2) can be calculated by multiplying θ(θ2) and θ(θ4) by the product of θ(θ2). Similarly, the fourth calibration parameter matrix Q(θ3) and the fifth calibration parameter matrix Q(θ4) can also be calculated.

[0041] Specifically, the first calibration parameter matrix Q(θ0) is calculated based on the implicit default probability matrix DP and the multi-year average migration matrix P; then the first preset value and the first calibration parameter inverse matrix Q(θ0) are calculated. -1 The second calibration parameter matrix Q(θ1) is calculated by multiplying the product of θ(θ1) and θ(θ2), and the third calibration parameter matrix Q(θ3), the fourth calibration parameter matrix Q(θ3) and the fifth calibration parameter matrix Q(θ4) are calculated by analogy;

[0042] S140. Calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0043] Among them, the calculation formula for calibrating the multi-year average migration matrix is: Q(t,t+1)=I+Q(θ)[PI], where Q(θ) is a calibration parameter matrix; Q(t,t+1) is the target credit default transfer matrix corresponding to the calibration parameter matrix Q(θ); the target credit default transfer matrix is ​​the credit default transfer matrix obtained after the calibration of the multi-year average migration matrix P. The credit default transfer matrix is ​​used to estimate the credit rating of corporate bonds and can improve the accuracy of credit rating estimation.

[0044] Specifically, taking the calibration parameter matrix Q(θ) corresponding to one of the years as an example, calculate the difference [PI] between the multi-year average migration matrix P and the unit matrix I, calculate the product of the difference [PI] and the calibration parameter matrix Q(θ), and then calculate the sum of the product and the unit matrix to obtain the target credit default transfer matrix Q(t,t+1) of the corresponding year.

[0045] It should be noted that, in this embodiment, an initial probability transfer matrix set is constructed based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume; weighted calculation is performed on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix; an implicit default probability matrix is ​​calculated based on the multi-year average migration matrix, and a calibration parameter matrix is ​​calculated based on the implicit default probability matrix and a preset calibration parameter inverse matrix; the multi-year average migration matrix is ​​calibrated based on the calibration parameter matrix to obtain a target credit default transfer matrix; through the above implementation, by introducing multi-dimensional data sources (such as initial credit rating, period rating situation and initial rating data volume, etc.), the impact of different factors on the credit of the enterprise is comprehensively considered, so as to truly reflect the fluctuation of the credit rating of the enterprise; by selecting the initial probability transfer matrices of different years for weighted calculation, the target credit default transfer matrix finally calibrated can be not only suitable for the prediction of rating changes of short-term bonds, but also can provide effective risk assessment for long-term bond investment; further, the use of the target credit default transfer matrix can more accurately predict the credit rating changes of corporate bonds.

[0046] Embodiment 2

[0047] Figure 2 A flow chart of a credit default transfer matrix calibration method provided in Example 2 of this application, refer to Figure 2 The method may be executed by a device for executing the method, and the device may be implemented by software and / or hardware. It should be noted that for the parts not described in detail in this embodiment, reference may be made to the descriptions of other embodiments. The method includes:

[0048] S210. Construct an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, interim rating situation, and initial rating data volume.

[0049] S220, performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix.

[0050] S230. Calculate an implicit default probability matrix based on the multi-year average migration matrix, and calculate a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix.

[0051] S240. Calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0052] S250: Obtain diagonal elements of the multi-year average migration matrix.

[0053] Among them, and the diagonal elements are denoted by p ii ; Diagonal element p ii That is, the elements on the diagonal of the multi-year average migration matrix P.

[0054] Specifically, obtain the elements p on the diagonal of the multi-year average migration matrix P ii .

[0055] S260. Calculate a risk-free probability transfer matrix corresponding to the multi-year average migration matrix based on the diagonal elements and the multi-year average migration matrix.

[0056] Among them, each element q in the risk-free probability transfer matrix Q ij The calculation formula is as follows:

[0057] q ij =p ij log(p ii ) / (p ii -1),(i≠j);

[0058] q ii =log(p ii );

[0059] Specifically, the diagonal elements p of the multi-year average migration matrix P are ii and the element p in the multi-year average migration matrix P ij Substitute into the above calculation formula to calculate, so as to obtain the elements q in the risk-free probability transfer matrix ij, and then the risk-free probability transfer matrix Q is obtained; the risk-free probability transfer matrix Q is subsequently used to input into the credit assessment model and jointly participate in the estimation of corporate bond credit ratings with the target credit default transfer matrix, thereby improving the accuracy of credit rating estimation.

[0060] Embodiment 3

[0061] Embodiment 3 of the present application provides a credit default transfer matrix calibration method, which is used to refine the "constructing an initial probability transfer matrix set based on credit rating information of enterprises with multiple years" in Embodiment 1. It should be noted that for the parts not described in detail in this embodiment, please refer to the description of other embodiments. The method includes:

[0062] S311. Calculate the probability of renewal, probability of default, repayment and other probabilities based on the rating situation during the period and the initial rating data volume.

[0063] Among them, the period rating situations include: renewal of rating, default, repayment and others; the credit rating information includes: initial credit rating, period rating situations and initial rating data volume.

[0064] Among them, the renewal rating refers to the situation that the credit rating of corporate bonds has not defaulted during the monitoring period and there is a valid rating; the default rating refers to the situation that the credit rating of corporate bonds has not defaulted during the monitoring period; the repayment and others refer to the situation that the credit rating of corporate bonds has no continuous rating record during the statistical period due to tracking, repayment, termination of rating or other factors except for default during the monitoring period. The renewal probability is the proportion of the credit rating of the renewal rating to the initial rating data volume; the default probability is the proportion of the credit rating of the default to the initial rating data volume; the repayment and others probability is the proportion of the repayment and others credit rating to the initial rating data volume.

[0065] Specifically, the renewal probability, default probability, repayment and other probabilities are calculated based on the rating situations during the period and the amount of initial rating data.

[0066] S312. Calculate the probability of transfer between ratings based on the probability of renewal, probability of default, repayment and other probabilities.

[0067] The calculation formula for the probability of transition between ratings is:

[0068] The probability of inter-rating transfer = the probability of renewal * the survival ratio / (1-payment and other probabilities); where the survival ratio = 1-default probability; the probability of inter-rating transfer is the element in the initial probability transfer matrix;

[0069] Specifically, the difference between 1 and the probability of default is calculated to obtain the survival ratio; then the product of the renewal probability and the survival ratio is calculated, and then the difference between 1 and the repayment and other probabilities is calculated; further, the ratio of the product and the difference is calculated to obtain a corresponding element in the initial probability transfer matrix, and the elements corresponding to each credit rating are calculated in the above manner.

[0070] S313. Obtain an initial probability transfer matrix of corresponding years based on the inter-rating transfer probability and the initial credit rating.

[0071] Among them, combining the inter-rating transfer probabilities corresponding to each initial credit rating can form an initial probability transfer matrix for the corresponding years.

[0072] Specifically, the inter-rating transfer probabilities corresponding to each initial credit rating are combined to form an initial probability transfer matrix of the corresponding years.

[0073] S314. Obtain an initial probability transfer matrix set based on the initial probability transfer matrices of multiple years.

[0074] Among them, the initial probability transfer matrix set is a set of initial probability transfer matrices corresponding to different years.

[0075] Specifically, the initial probability transfer matrices corresponding to different years are calculated through the above steps, and then the initial probability transfer matrices corresponding to different years are aggregated to obtain an initial probability transfer matrix set.

[0076] S320, performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix.

[0077] S330. Calculate an implicit default probability matrix based on the multi-year average migration matrix, and calculate a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix.

[0078] S340. Calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0079] Embodiment 4

[0080] Embodiment 4 of the present application provides a credit default transfer matrix calibration method, which is used to refine the "weighted calculation of each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix" in Embodiment 1. It should be noted that for the parts not described in detail in this embodiment, please refer to the description of other embodiments. The method includes:

[0081] S410. Construct an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, interim rating situation, and initial rating data volume.

[0082] S421. Merge the data corresponding to the preset rating set in the initial probability transfer matrix to obtain a simplified probability transfer matrix.

[0083] It should be noted that in the market, the number of issuers with low credit ratings is limited, so the data in these low credit rating categories are often scarce. This data scarcity may affect the reliability of the analysis. In order to solve this problem, we merge the data in the low credit rating category; the preset rating in this embodiment is also a low credit rating. In this embodiment, the set of credit ratings below the BBB credit rating is used as the preset rating set; the merger is to add the matrix elements corresponding to each credit rating in the preset rating set in the initial probability transfer matrix and divide it by the number of credit ratings in the preset rating set to obtain the corresponding merger result; the simplified probability transfer matrix is ​​also the initial probability transfer matrix after the data merger is completed.

[0084] Specifically, taking a column of data corresponding to a credit rating in the initial probability transfer matrix as an example, the data in the column of data corresponding to the preset rating set is determined, and then the data corresponding to the preset rating set is merged to obtain a simplified probability transfer matrix.

[0085] S422, performing weighted calculation on the simplified probability transfer matrices corresponding to the initial probability transfer matrices in the initial probability transfer matrix set to obtain a multi-year average migration matrix.

[0086] Among them, each simplified probability transfer matrix has a corresponding initial probability transfer matrix, and the simplified probability transfer matrix is ​​generated based on the corresponding initial probability transfer matrix; the multi-year average migration matrix is ​​a matrix obtained by weighted calculation of the simplified probability transfer matrices corresponding to multiple years.

[0087] Specifically, the simplified probability transfer matrices corresponding to multiple years are weighted and calculated to obtain a multi-year average migration matrix.

[0088] S430. Calculate an implicit default probability matrix based on the multi-year average migration matrix, and calculate a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix.

[0089] S440. Calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0090] Embodiment 5

[0091] Embodiment 5 of the present application provides a credit default transfer matrix calibration method, which is used to refine the "weighted calculation of the simplified probability transfer matrix corresponding to each of the initial probability transfer matrices in the initial probability transfer matrix set to obtain a multi-year average migration matrix" in Embodiment 4. It should be noted that for the parts not described in detail in this embodiment, please refer to the description of other embodiments. The method includes:

[0092] S510. Construct an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume.

[0093] S521. Merge the data corresponding to the preset rating set in the initial probability transfer matrix to obtain a simplified probability transfer matrix.

[0094] S5221. Match corresponding matrix weights for each simplified probability transfer matrix.

[0095] Among them, this embodiment specifically adopts a simplified probability transfer matrix for a five-year period and a simplified probability transfer matrix for a ten-year period; the matrix weight of the simplified probability transfer matrix matching for a five-year period may be 0.5, and the matrix weight of the simplified probability transfer matrix matching for a ten-year period is also 0.5; in other embodiments, the matrix weight of the simplified probability transfer matrix matching for a five-year period may be 0.4. At this time, the matrix weight of the simplified probability transfer matrix matching for a ten-year period is 0.6; and so on.

[0096] Specifically, each simplified probability transfer matrix is ​​matched with a corresponding matrix weight.

[0097] S5222. Perform weighted calculation on each simplified probability transfer matrix based on the weight of each matrix to obtain a multi-year average migration matrix.

[0098] Specifically, the product of each simplified probability transfer matrix and the corresponding matrix weight is calculated, and then the corresponding products of each simplified probability transfer matrix are added to calculate the multi-year average migration matrix.

[0099] S530. Calculate an implicit default probability matrix based on the multi-year average migration matrix, and calculate a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix.

[0100] S540. Calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0101] Embodiment 6

[0102] Embodiment 6 of the present application provides a credit default transfer matrix calibration method, which is used to refine the "calculating an implicit default probability matrix based on the multi-year average migration matrix" in Embodiment 1. It should be noted that for the parts not described in detail in this embodiment, reference can be made to the descriptions of other embodiments. The method includes:

[0103] S610. Construct an initial probability transfer matrix set based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, interim rating situation, and initial rating data volume.

[0104] S620, performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix.

[0105] S631. Construct a bond interest rate matrix based on the treasury bond interest rates and credit bond interest rates corresponding to the multi-year average migration matrix.

[0106] Among them, the multi-year average migration matrix has the corresponding treasury bond interest rate trea and credit bond interest rate in terms of years; the bond interest rate matrix can be obtained by combining the treasury bond interest rate trea and credit bond interest rate of different credit ratings.

[0107] Specifically, the treasury bond interest rates trea and credit bond interest rates of different credit ratings corresponding to the multi-year average migration matrix are combined to construct a bond interest rate matrix.

[0108] S632. Calculate a bond price matrix based on the bond interest rate matrix.

[0109] The bond price matrix is ​​calculated based on the bond price matrix, and each element in the bond price matrix is ​​calculated based on the corresponding element in the bond interest rate matrix; and the element in the bond interest rate matrix is ​​recorded as r(t), and the corresponding element in the bond price matrix is ​​recorded as v(t), and the calculation formula between v(t) and r(t) is: v(t) = e -r(t)t ; Wherein, t is the number of years; for example, the t corresponding to a five-year period is 5.

[0110] Specifically, each corresponding element v(t) in the bond price matrix is ​​calculated according to each element r(t) in the bond interest rate matrix and the corresponding term t, thereby obtaining the bond price matrix.

[0111] S633. Calculate a survival rate matrix based on the bond price matrix, the multi-year average migration matrix, and a preset bond return rate.

[0112] Among them, the bond market has its corresponding bond return rate δ. For example, the bond return rate δ is 0.4; each element Q in the survival rate matrix t The calculation method is:

[0113]

[0114] Specifically, the corresponding element v(t) in the bond price matrix, the multi-year average migration matrix P, and the bond return rate δ are substituted into the above formula for calculation, so as to obtain the corresponding element of the survival rate matrix, and then obtain the survival rate matrix.

[0115] S634. Calculate an implicit default probability matrix based on a preset identity matrix and the survival rate matrix.

[0116] Among them, the implicit default probability matrix is ​​the difference between the identity matrix and the survival rate matrix.

[0117] Specifically, the difference between the identity matrix and the survival rate matrix is ​​calculated to obtain the implicit default probability matrix.

[0118] S635. Calculate a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix.

[0119] S640. Calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0120] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0121] Embodiment 7

[0122] Based on the same inventive concept, this embodiment also provides a credit default transfer matrix calibration device for implementing the credit default transfer matrix calibration method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more credit default transfer matrix calibration device embodiments provided below can refer to the limitations of the credit default transfer matrix calibration method above, and will not be repeated here.

[0123] In this embodiment, Figure 3 As shown, a credit default transfer matrix calibration device is provided, comprising:

[0124] A matrix set construction module, used to construct an initial probability transfer matrix set based on credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume;

[0125] A weighted calculation module, used for performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix;

[0126] A calibration matrix calculation module, used for calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix;

[0127] A matrix calibration module is used to calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

[0128] Embodiment 8

[0129] Based on the same inventive concept, this embodiment also provides a credit default transfer matrix calibration device for implementing the credit default transfer matrix calibration method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more credit default transfer matrix calibration device embodiments provided below can refer to the limitations of the credit default transfer matrix calibration method above, and will not be repeated here.

[0130] In this embodiment, Figure 4 As shown, a credit default transfer matrix calibration device is provided, comprising:

[0131] A matrix set construction module, used to construct an initial probability transfer matrix set based on credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume;

[0132] A weighted calculation module, used for performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix;

[0133] A calibration matrix calculation module, used for calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix;

[0134] A matrix calibration module, used for calibrating the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix;

[0135] An element acquisition module, used for acquiring diagonal elements of the multi-year average migration matrix;

[0136] A matrix calculation module is used to calculate a risk-free probability transfer matrix corresponding to the multi-year average migration matrix based on the diagonal elements and the multi-year average migration matrix.

[0137] Each module in the above-mentioned credit default transfer matrix calibration device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0138] It should be noted that, in this embodiment, an initial probability transfer matrix set is constructed based on the credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume; weighted calculation is performed on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix; an implicit default probability matrix is ​​calculated based on the multi-year average migration matrix, and a calibration parameter matrix is ​​calculated based on the implicit default probability matrix and a preset calibration parameter inverse matrix; the multi-year average migration matrix is ​​calibrated based on the calibration parameter matrix to obtain a target credit default transfer matrix; through the above implementation, by introducing multi-dimensional data sources (such as initial credit rating, period rating situation and initial rating data volume, etc.), the impact of different factors on the credit of the enterprise is comprehensively considered, so as to truly reflect the fluctuation of the credit rating of the enterprise; by selecting the initial probability transfer matrices of different years for weighted calculation, the target credit default transfer matrix finally calibrated can be not only suitable for the prediction of rating changes of short-term bonds, but also can provide effective risk assessment for long-term bond investment; further, the use of the target credit default transfer matrix can more accurately predict the credit rating changes of corporate bonds.

[0139] Embodiment 9

[0140] In this embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a credit default transfer matrix calibration method is implemented.

[0141] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present disclosure, and does not constitute a limitation on the computer device to which the scheme of the present disclosure is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0142] Embodiment 10

[0143] In this embodiment, a computer readable storage medium is provided. Figure 6 As shown, a computer program is stored thereon, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0144] Embodiment 11

[0145] In this embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0147] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided by the present disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided by the present disclosure may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited thereto. The processor involved in each embodiment provided by the present disclosure may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited thereto.

[0148] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0149] The above-described embodiments only express several implementation methods of the present disclosure, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present disclosure, and these all belong to the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the attached claims.

Claims

1. A credit default transfer matrix calibration method, characterized in that: include: Constructing an initial probability transfer matrix set based on credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume; Performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix; Calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix; The multi-year average migration matrix is ​​calibrated based on the calibration parameter matrix to obtain a target credit default transfer matrix.

2. The method according to claim 1, characterized in that: Also includes: Obtaining the diagonal elements of the multi-year average migration matrix; A risk-free probability transfer matrix corresponding to the multi-year average migration matrix is ​​calculated based on the diagonal elements and the multi-year average migration matrix.

3. The method according to claim 1, characterized in that The rating situations during the period include: renewal, default, repayment and others; the initial probability transfer matrix set constructed based on the credit rating information of enterprises with multiple years includes: Calculate the probability of renewal, probability of default, repayment and other probabilities based on the rating situations during the period and the amount of initial rating data; Calculate the probability of transfer between ratings based on the probability of renewal, probability of default, and repayment and other probabilities; Based on the inter-rating transfer probability and the initial credit rating, an initial probability transfer matrix of the corresponding years is obtained; An initial probability transfer matrix set is obtained based on the initial probability transfer matrices of multiple years.

4. The method according to claim 1, characterized in that: The weighted calculation of each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix includes: Merging the data corresponding to the preset rating set in the initial probability transfer matrix to obtain a simplified probability transfer matrix; The simplified probability transfer matrix corresponding to each of the initial probability transfer matrices in the initial probability transfer matrix set is weightedly calculated to obtain a multi-year average migration matrix.

5. The method according to claim 4, characterized in that The weighted calculation of the simplified probability transfer matrix corresponding to each of the initial probability transfer matrices in the initial probability transfer matrix set to obtain a multi-year average migration matrix includes: Matching corresponding matrix weights for each of the simplified probability transfer matrices; The simplified probability transfer matrices are weightedly calculated based on the weights of the matrices to obtain a multi-year average migration matrix.

6. The method according to claim 1, characterized in that The step of calculating an implicit default probability matrix based on the multi-year average migration matrix comprises: Constructing a bond interest rate matrix based on the government bond interest rates and credit bond interest rates corresponding to the multi-year average migration matrix; Calculate a bond price matrix based on the bond interest rate matrix; Calculating a survival rate matrix based on the bond price matrix, the multi-year average migration matrix, and a preset bond return rate; The implicit default probability matrix is ​​calculated based on the preset identity matrix and the survival rate matrix.

7. A credit default transfer matrix calibration device, characterized in that: The device comprises: A matrix set construction module, used to construct an initial probability transfer matrix set based on credit rating information of enterprises of multiple years, wherein the credit rating information includes: initial credit rating, period rating situation and initial rating data volume; A weighted calculation module, used for performing weighted calculation on each initial probability transfer matrix in the initial probability transfer matrix set to obtain a multi-year average migration matrix; A calibration matrix calculation module, used for calculating an implicit default probability matrix based on the multi-year average migration matrix, and calculating a calibration parameter matrix based on the implicit default probability matrix and a preset calibration parameter inverse matrix; A matrix calibration module is used to calibrate the multi-year average migration matrix based on the calibration parameter matrix to obtain a target credit default transfer matrix.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images