CN116664289A

CN116664289A - Service information prediction method and device, storage medium and electronic device

Info

Publication number: CN116664289A
Application number: CN202310791479.4A
Authority: CN
Inventors: 邱开睿
Original assignee: Industrial Consumer Finance Co Ltd
Current assignee: Industrial Consumer Finance Co Ltd
Priority date: 2023-06-29
Filing date: 2023-06-29
Publication date: 2023-08-29

Abstract

The application discloses a method and a device for predicting service information, a storage medium and an electronic device, wherein the method comprises the following steps: determining a target account age stage to which a designated account age of a target credit service belongs, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, and the N account ages of the target credit resource are divided into a plurality of account age stages; determining overdue rate parameters of each reference account age in a group of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the group of reference account ages comprises (n+1) th account ages of target credit resources issued from a 1 st time period to a (m-1) th time period; and predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is the overdue rate prediction mode corresponding to the target account age stage.

Description

Service information prediction method and device, storage medium and electronic device

Technical Field

The present application relates to the field of the internet, and in particular, to a method and apparatus for predicting service information, a storage medium, and an electronic apparatus.

Background

With the development of artificial intelligence algorithms and big data technologies, during credit risk management, data driving is usually used as a gripper to perform full-flow and multidirectional credit risk management and control, and the data technologies and the intelligent algorithms are also applied to aspects of risk measurement, prediction, efficiency improvement and the like. For credit businesses, asset quality monitoring and prediction is an important part of risk management and control, and a common risk monitoring and prediction method is account age analysis (i.e., account age method). As a standard for evaluating the quality of credit assets, the account age analysis method has the characteristics of objectivity, intuitiveness and comprehensiveness, and the account age curve dynamically describes the migration condition of the overdue rate of the credit asset in the whole life cycle.

At present, when predicting service information of credit service, a common prediction mode is an average acceleration method, that is, an acceleration rate of each of the accounts for periods (account age) is calculated first (that is, an acceleration rate of contract overdue rate), then the average overdue rate of each of the accounts for periods is calculated, and the average value is assigned to a corresponding overdue number of the unexplained loan, so as to calculate a final overdue rate performance.

However, the accuracy of the average acceleration method depends on the early-stage account age performance, and when the current overdue rate (i.e., the contract overdue rate) is high, the predicted overdue rates may be higher than the actual result, and when the current overdue rate is low, the predicted overdue rates may be lower than the actual result, resulting in lower accuracy of overdue rate prediction. Therefore, the prediction method of the service information in the related art has the problem of low accuracy of the service information prediction.

Disclosure of Invention

The embodiment of the application provides a service information prediction method and device, a storage medium and an electronic device, which at least solve the problem that the service information prediction accuracy is low in the service information prediction method in the related technology.

According to an aspect of an embodiment of the present application, there is provided a method for predicting service information, including: determining a target account age stage to which a designated account age of a target credit service belongs, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N; determining overdue rate parameters of each reference account age in a set of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the set of reference account ages comprises (n+1) th account ages of the target credit resource issued from a 1 st time period to a (m-1) th time period; and predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

According to another aspect of the embodiment of the present application, there is also provided a service information prediction apparatus, including: a first determining unit, configured to determine a target account age stage to which a designated account age of a target credit service belongs, where the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N; a second determining unit, configured to determine an overdue rate parameter of each reference account age in a set of reference account ages matched with a target account age, where the target account age is a next account age of a designated account age, and the set of reference account ages includes (n+1) th account age of the target credit resource issued in a 1 st to (m-1) th period; the prediction unit is used for predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

In one exemplary embodiment, the prediction unit includes: and the prediction module is used for predicting the overdue rate corresponding to the target account age according to the overdue rate parameter corresponding to each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is the overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

In one exemplary embodiment, the set of reference account age is an (n+1) th account age of the target credit resource issued from a 1 st period to an (m-1) th period, in a case where the target account age stage is a first one of the plurality of account age stages; the prediction module includes: and the first determining submodule is used for determining the average value of the overdue rate corresponding to each reference account age as the target overdue rate.

In one exemplary embodiment, the set of reference account ages includes an (n+1) th account age of the target credit resource issued by the designated account age and time 1 to (m-1) th time period, in a case where the target account age phase is a last account age phase of the plurality of account age phases; the prediction module includes: the second determining submodule is used for determining an average value of the overdue rate acceleration corresponding to the (n+1) th account age of the target credit resource issued in the 1 st time period to the (m-1) th time period to obtain an average value of the overdue rate acceleration, and determining a standard value of the overdue rate acceleration corresponding to the (n+1) th account age of the target credit resource issued in the 1 st time period to the (m-1) th time period to obtain a standard value of the overdue rate acceleration; the third determining submodule is used for determining a target normal distribution function according to the overdue rate acceleration mean value and the overdue rate acceleration standard value, wherein the overdue rate acceleration mean value is a position parameter of the target normal distribution function, and the overdue rate acceleration standard value is a shape parameter of the target normal distribution function; a fourth determining submodule, configured to determine a function value corresponding to the target normal distribution function and a random value, and determine an estimated overdue rate acceleration corresponding to the target account age; and a fifth determining submodule, configured to determine the target overdue rate according to a product of the overdue rate corresponding to the specified account age and the overdue rate estimated acceleration corresponding to the target account age.

In one exemplary embodiment, in the case where the target account age stage is an account age stage other than a first account age stage and a last account age stage of the plurality of account age stages, the set of reference account ages includes an (n+1) th account age of the target credit resource issued by the designated account age stage and a 1 st to (m-1) th period; the prediction module includes: the aggregation sub-module is used for carrying out aggregation processing on the overdue rate acceleration of each reference account age by adopting a designated aggregation mode under the condition that the target account age stage is an account age stage except a first account age stage and a last account age stage in the plurality of account age stages, so as to obtain the overdue rate estimated acceleration of the target account age; and the sixth determining submodule is used for determining the product of the overdue rate of the appointed account age and the overdue rate estimated acceleration of the target account age as the target overdue rate.

In an exemplary embodiment, the apparatus further comprises: and the third determining unit is used for determining the ratio between the overdue rate of each reference account age and the overdue rate of the last account age of each reference account age as the overdue rate acceleration of each reference account age.

In an exemplary embodiment, the first determining unit includes: the first determining module is used for determining the overdue rate acceleration of the target credit business in the appointed account age according to the overdue rate corresponding to the last account age of the appointed account age and the overdue rate corresponding to the appointed account age; and the second determining module is used for determining the target account age stage to which the appointed account age belongs according to the overdue rate acceleration of the target credit business in the appointed account age.

In one exemplary embodiment, the second determining module includes: a seventh determining submodule, configured to determine, when the overdue rate acceleration of the target credit service at the specified account age is greater than or equal to a first acceleration threshold, that the target account age stage to which the specified account age belongs is a first account age stage; an eighth determining submodule, configured to determine, when the overdue rate acceleration of the target credit service at the specified account age is greater than or equal to a second acceleration threshold and less than the first acceleration threshold, that the target account age stage to which the specified account age belongs is a second account age stage; a ninth determining submodule, configured to determine, when the overdue rate acceleration of the target credit service at the specified account age is less than a second acceleration threshold, that the target account age stage to which the specified account age belongs is a third account age stage; the first speed increasing threshold is larger than the second speed increasing threshold, and all account ages of the target credit business are divided into the first account age stage, the second account age stage and the third account age stage in sequence from front to back.

According to a further aspect of the embodiments of the present application, there is also provided a computer-readable storage medium having stored therein a computer program, wherein the computer program is configured to execute the above-described service information prediction method when run.

According to still another aspect of the embodiments of the present application, there is further provided an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the method for predicting service information according to the above-mentioned method.

In the embodiment of the application, a mode of dividing the whole account age period of a credit service into different account age stages and predicting the overdue rate of account ages belonging to the different account age stages in different prediction modes is adopted, and the target account age stage to which a designated account age of a target credit service belongs is determined, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N; determining overdue rate parameters of each reference account age in a set of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the set of reference account ages comprises (n+1) th account ages of the target credit resource issued from a 1 st time period to a (m-1) th time period; according to the overdue rate parameter of each reference account age, the overdue rate of the target account age is predicted according to a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is the overdue rate prediction mode corresponding to the target account age stage, different account age stages correspond to different overdue rate prediction modes, and the corresponding overdue rate prediction modes are respectively set for different account age stages of the account age curve by adopting a divide-and-conquer mode, so that the influence of the accuracy of the overdue rate prediction on the accuracy of the later overdue rate prediction can be achieved, the technical effect of improving the accuracy of the service information prediction can be achieved, and the problem that the service information prediction method in the related technology has low accuracy of the service information prediction is solved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.

FIG. 1 is a schematic diagram of a hardware environment of an alternative business information prediction method according to an embodiment of the present application;

FIG. 2 is a flow chart of an alternative method of predicting business information in accordance with an embodiment of the present application;

FIG. 3 is a schematic illustration of an alternative account age curve according to an embodiment of the present application;

FIG. 4 is a schematic diagram of an alternative correlation of current overdue rate to next overdue rate according to an embodiment of the application;

FIG. 5 is a schematic diagram of an alternative normal distribution according to an embodiment of the application;

FIG. 6 is a flow chart of another alternative method of predicting business information in accordance with an embodiment of the present application;

Fig. 7 is a block diagram showing the construction of an alternative service information prediction apparatus according to an embodiment of the present application;

fig. 8 is a block diagram of an alternative electronic device according to an embodiment of the application.

Detailed Description

In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

According to an aspect of an embodiment of the present application, there is provided a method for predicting service information. The method for predicting service information may be in a hardware environment including the terminal device 102 and the server 104 as shown in fig. 1. As shown in fig. 1, the server 104 is connected to the terminal device 102 through a network, and may be used to provide services (such as application services and the like) for a terminal or a client installed on the terminal, a database may be set on the server or independent of the server, for providing data storage services for the server 104, and cloud computing and/or edge computing services may be configured on the server or independent of the server, for providing data computing services for the server 104.

The network may include, but is not limited to, at least one of: wired network, wireless network. The wired network may include, but is not limited to, at least one of: a wide area network, a metropolitan area network, a local area network, and the wireless network may include, but is not limited to, at least one of: WIFI, bluetooth. The terminal device 102 may be, but is not limited to, a PC, a mobile phone, a tablet computer, a smart home device, etc.

The method for predicting service information according to the embodiment of the present application may be performed by the server 104, may be performed by the terminal device 102, or may be performed by both the server 104 and the terminal device 102. The method for predicting the service information performed by the terminal device 102 according to the embodiment of the present application may be performed by a client installed thereon.

Taking the example that the server 104 performs the method for predicting service information in this embodiment, fig. 2 is a schematic flow chart of an alternative method for predicting service information according to an embodiment of the present application, as shown in fig. 2, the flow of the method may include the following steps:

step S202, determining a target account age stage to which a designated account age of a target credit service belongs, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N.

The prediction method of business information in the present embodiment can apply a scenario of predicting the expiration rate in the account age of a credit resource (i.e., a credit asset) corresponding to a specified credit business issued in one period. The specified credit service may be a loan service applied for a transaction by the user to a credit agency, and the transaction result may be determined based on the credit points of the user. The account age of the credit resource may refer to the number of specified time periods that the credit resource has passed from the paying-out to the current time, and the specified time period may be one day, half month or one month, which may be flexibly set as needed.

Illustratively, the correspondence between the account age (Month of Book) and the expiration rate may be shown in FIG. 3 as an account age curve (i.e., a Vintage curve), wherein the abscissa of the curve is the account age of the credit asset and the ordinate is the contract expiration rate (the ordinate may range from 0-5%) of the credit asset for the account age, and the ideal account age curve is an increasing upward convex function with sufficient exposure of risk after a certain period of time, the curve gradually stabilizes.

In the related art, when the overdue rate is predicted, the average overdue rate increasing mean value can be determined through an average increasing method, and the overdue rate of the next account age is predicted based on the overdue rate increasing mean value combined with the overdue rate of the previous account age. However, since the accuracy of the prediction depends on the early-stage account age performance, and there is no difference in the rate of expiration of the account age, the above-mentioned manner of predicting the rate of expiration has a problem in that the prediction of the rate of expiration is inaccurate. Here, the accuracy of the average acceleration rate prediction result is greatly dependent on the prediction of the early-stage overdue rate, and if the early-stage overdue rate is overestimated, the late-stage overdue rate is overestimated, and the deviation is larger.

Meanwhile, since the change rules of overdue rate data in different periods are not considered, for example, in the occurrence of sudden conditions, in different periods of economic development and the like, if a more accurate prediction result is expected, abnormal data is required to be removed or similar macroscopic economic environment is matched for calculation, and high-quality and enough large data samples are required, in practical application, after the limiting condition is added, the data amount for model training is often insufficient, and the robustness of the model is reduced, so that the applicability of an average acceleration method is poor.

In addition, the overdue rate prediction of the account age can be performed based on expert experience methods, and the core of the method is to predict the contract overdue rate acceleration of loans in each period, wherein the expert experience methods are to predict the trend of the contract overdue rate by combining the internal management and assessment of an expert and the external environment of an institution. Meanwhile, the comprehensive judgment can be carried out by combining with an approval strategy and a wind control tool in the mechanism, so that the final overdue risk is determined. However, the expert experience method excessively depends on manual experience judgment, batch prediction cannot be realized, the efficiency is low, the prediction cost is high, and the economical efficiency is poor.

In order to at least partially solve the above technical problems, in terms of economy, accuracy, applicability and the like of the overdue rate prediction, in this embodiment, a separate method is adopted to set a corresponding overdue rate prediction method for different account age stages of the account age curve, that is, the account age number (or the account age curve) is divided into different account age stages, so as to adapt to the overdue rate of different account ages and the change of the overdue rate acceleration, and a corresponding prediction method is set based on the characteristics of the overdue rate change of different account age stages, so that the overdue rate of the account ages belonging to different account age stages is predicted by adopting different prediction methods, the prediction result is more reasonable, meanwhile, because the overdue rate prediction is completely based on data driving, the manual intervention (that is, the process of manual intervention is reduced), the judging process is greatly simplified, and the batch measurement and calculation can be performed according to the credit performance of the product (corresponding to the service) per se, so that the labor cost is saved.

In addition, the prediction mode of the overdue rate has strong generalization capability, when predicting, if the loan (namely, credit business) has historical risk expression, the self data can be used for recursive prediction, and if the historical expression is insufficient, the credit product account age which is similar to the characteristic and has longer expression period can be used for calculation.

For the target credit business, the credit resource corresponding to the target credit business is the target credit resource (or referred to as the target credit asset), the target credit resource has N account age periods (i.e. the repayment period of the target credit resource is divided into N periods, N is a positive integer greater than or equal to 1), the N account age periods may be divided into a plurality of account age periods, the manner of division may be manual division based on experience values, division may be based on analysis of historical data, and the plurality of account age periods may be at least two account age periods, for example, three account age periods, or other number of account age periods, which is not limited in this embodiment.

For the account age to be predicted, that is, the target account age, the account age to which the target account age belongs may be determined based on the account age stage to which the previous account age belongs, wherein the account age stage to which the one account age belongs may be determined based on the overdue rate, the overdue rate acceleration rate, the overdue rate of the previous account age, and the like, and further, may be determined based on the number of the target account ages, that is, the correspondence between the preset account age and the account age stage, and the account age stage corresponding to the target account age may be determined based on the correspondence between the preset account age and the account age stage. Taking the account age stage to which the account age to be predicted belongs as an example based on the account age stage to which the previous account age belongs, the target account age stage to which the designated account age of the target credit service belongs may be first determined, where the designated account age is the previous account age of the target account age.

Here, the credit resources of the same credit service may be issued in a time period, for example, once per month, then, each time the credit resources issued are issued in N account periods, and the statement of the previous account period, the statement of the next account period, or the like of one account period is in terms of the credit resources issued in the same time period, for the target account period, the nth account period of the target credit resources issued in the mth time period and corresponding to the target credit service, m, N are positive integers greater than or equal to 1, and N is less than or equal to N.

Step S204, determining overdue rate parameters of each reference account age in a group of reference account ages matched with the target account age, wherein the group of reference account ages comprises (n+1) th account age of the target credit resource issued from the 1 st time period to the (m-1) th time period.

For the target account age (the next account age of the designated account age), the overdue rate prediction may be referred to by the overdue rate parameter (e.g., overdue rate acceleration, etc.) of the (n+1) th account age of the target credit resource corresponding to the target credit service issued in other period, and thus the overdue rate parameter of each reference account age of a set of reference account ages matching the target account age may be determined, the set of reference account ages including the (n+1) th account age of the target credit resource issued in the 1 st to (m-1) th periods. Here, the target account age is the (n+1) th account age of the target credit resource issued n times, and the target credit resource generally refers to a credit service corresponding to the target credit service.

For example, every month (i.e., every period) of 1 month to 5 months of a certain year, a credit asset of a certain credit business is issued. The repayment period of the credit asset is 9 periods (namely, 9 months), and when the overdue rate of the 5 th period of the credit asset issued by 4 months of the year is predicted, the group of reference account ages corresponding to the overdue rate is the 5 th period of the credit asset issued by 1 month to 3 months of the year.

Step S206, predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

For different account age stages, the corresponding overdue rate prediction modes can be set respectively, the different account age stages correspond to the different overdue rate prediction modes, and the basis for setting the overdue rate prediction modes can be the overdue rate in the corresponding account age stage or the change rule of the overdue rate acceleration, wherein the overdue rate prediction mode corresponding to the target account age stage is the target prediction mode.

For the target account age, the overdue rate of the target account age can be predicted according to the overdue rate parameter of each reference account age in a target prediction mode, and the obtained target overdue rate is based on the predicted overdue rate, namely, the predicted value of the overdue rate. Here, the overdue rate parameter used in the corresponding overdue rate prediction method may be different for different account periods, for example, at least one of the overdue rate and the overdue rate acceleration. When predicting the overdue rate of the target account age, the overdue rate acceleration of the target account age can be predicted according to a target prediction mode, and then the target overdue rate is determined based on the predicted overdue rate acceleration, or the overdue rate of the target account age can be predicted directly according to the target prediction mode. This is not limited in this embodiment.

Through the steps S202 to S206, determining a target account age stage to which a designated account age of the target credit service belongs, where the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N; determining overdue rate parameters of each reference account age in a group of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the group of reference account ages comprises (n+1) th account ages of target credit resources issued from a 1 st time period to a (m-1) th time period; according to the overdue rate parameter of each reference account age, the overdue rate of the target account age is predicted according to a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is the overdue rate prediction mode corresponding to the target account age stage, different account age stages correspond to different overdue rate prediction modes, the problem that the service information prediction accuracy is low in the service information prediction method in the related technology is solved, and the service information prediction accuracy is improved.

In an exemplary embodiment, predicting the expiration rate of the target account age according to the expiration rate parameter of each reference account age in a target prediction mode to obtain a target expiration rate, including:

s11, when the target account age stage is the first account age stage in the plurality of account age stages, determining the average value of the overdue rates corresponding to each reference account age as the target overdue rate.

For the first account age stage of the plurality of account age stages, which is the early stage of account age growth, the credit risk is not fully represented and is easily influenced by extreme values, and for model robustness, the prediction of the growth rate (namely, the acceleration rate) is not performed in the stage, but the contract expiration rate of a few months is directly used for simple average or weighted average, and in addition, the abnormal value which is beyond 2 times of standard deviation can be eliminated. Alternatively, the average of the overdue rates corresponding to each reference account age may be determined as the target overdue rate, and the average may be an average or a weighted average, in which case the overdue rate parameter of each reference account age is the overdue rate of each reference account age.

For the first account age stage (for example, the first stage, stage I), the overdue rate of the target account age may be predicted by using a first prediction model (for example, stage I model) according to the overdue rate of each reference account age, so as to obtain a target overdue rate, that is, the overdue rate of each reference account age is input into the first prediction model, so as to obtain the target overdue rate output by the first prediction model. The first predictive model is a predictive model corresponding to the first accounting period.

Here, as shown in fig. 4, stage i is usually in the early stage of the account age curve, the data is represented by mean value and variance of contract overdue rate, mean variance and standard deviation of acceleration are large, and at this time, the account age trend deviation is estimated to be too large by adopting acceleration transfer, so that the recent contract overdue rate smoothing process can be directly adopted for prediction.

According to the embodiment, in the early stage of the increase of the account age, the accuracy of the overdue prediction can be improved by taking the average or weighted average of the overdue rate as the predicted overdue rate.

s21, under the condition that the target account age stage is the last account age stage in the plurality of account age stages, determining the average value of the overdue rate acceleration of each reference account age to obtain the average value of the overdue rate acceleration, and determining the standard value of the overdue rate acceleration of each reference account age to obtain the standard value of the overdue rate acceleration;

s22, determining a target normal distribution function according to the overdue rate acceleration average value and an overdue rate acceleration standard value, wherein the overdue rate acceleration average value is a position parameter of the target normal distribution function, the overdue rate acceleration standard value is a shape parameter of the target normal distribution function, and the target normal distribution function is a normal distribution function between overdue rate and overdue rate acceleration;

S23, determining a function value corresponding to a random expected rate in a target normal distribution function and overdue rate range as overdue rate estimated acceleration of a target account age, wherein the overdue rate range comprises overdue rates of each reference account age;

s24, determining the product of the overdue rate of the appointed account age and the overdue rate estimated acceleration of the target account age as the target overdue rate.

For the last account age stage in the plurality of account age stages, the stage is usually at the end of the account age curve, the curve gradually reaches a stable state, and the data is represented by a high average value of contract overdue rate, but the variance is basically approximate to 0. Since curve fluctuations at this stage can be interpreted by white noise, it can be assumed that the rate of rise at this stage follows a normal distribution, and risk inference is performed. Alternatively, the normal distribution function corresponding to the target account age, that is, the target normal distribution function, which is a normal distribution function between the overdue rate and the overdue rate acceleration rate may be first determined according to the overdue rate acceleration rate of each reference account age; and then, determining the function value corresponding to the target normal distribution function and a random expected rate as the overdue rate estimated acceleration of the target account age. In this case, the overdue rate parameter of each reference account age is a rate of increase in the overdue rate of each reference account age. Here, the above-described one random expectation rate is one random number within a range of expiration rates including the expiration rate of each reference account age.

After the estimated acceleration of the overdue rate of the target account age is obtained, the product of the overdue rate of the appointed account age and the estimated acceleration of the overdue rate of the target account age can be determined as the target overdue rate. For example, the n+1st phase pre-estimated contract expiration rate may be calculated according to equation (1):

odr' _m,n+1 ＝odr _m,n ×inr' _m,n+1 (1)

wherein odr' _m,n+1 Predicted value of contract overdue rate of m month input loan at n+1st period odr _m,n Indicating the overdue rate of the n-th contract of the m-month put loan, inr' _m,n+1 Indicating the estimated acceleration of the overdue rate of the contract in the n+1st period of the m month put loan.

When determining the target normal distribution function, the position parameter of the target normal distribution function and the shape parameter of the target normal distribution function can be respectively determined: the average value of the overdue rate acceleration of each reference account age can be determined, the average value of the overdue rate acceleration is obtained, the standard value of the overdue rate acceleration of each reference account age is determined, and the standard value of the overdue rate acceleration is obtained, wherein the average value of the overdue rate acceleration is the position parameter of the target normal distribution function, and the standard value of the overdue rate acceleration is the shape parameter of the target normal distribution function.

For the last account age stage (for example, the third stage, stage iii), the overdue rate of the target account age can be predicted by using a third prediction model (for example, the stage iii model) according to the overdue rate acceleration of each reference account age, so as to obtain a target overdue rate, that is, the overdue rate acceleration of each reference account age is input into the third prediction model, so as to obtain the target overdue rate output by the third prediction model. The third predictive model is a predictive model corresponding to the last ledger stage.

For example, stage III tends to be at the end of the ledger's age curve, which has been smoothed, the data is shown to be high in the mean of contract expiration rates, but the variance is substantially approaching 0. Although the curve fluctuates slightly, the curve fluctuation can be interpreted by white noise through the detection of white noise (Ljung-Box), and can be based on inr' _m,n ～N(μ _n ,σ _n ) Assume that the phase III is modeled, where inr' _m,n The contract overdue rate of the loan in the n period is estimated to be increased for m months. Mu (mu) _n ，σ _n An example of a normal distribution may be shown in FIG. 5, with the mean and standard deviation of the acceleration rate of the n-term contract for each month of the loan being issued.

According to the embodiment, at the end of the account age curve, the corresponding normal distribution function is determined, the predicted overdue rate acceleration is determined based on the random expected rate, the predicted overdue rate acceleration can be matched with the curve fluctuation rule, and the overdue prediction accuracy is improved.

s31, under the condition that the target account age stage is an account age stage except a first account age stage and a last account age stage in a plurality of account age stages, adopting a designated aggregation mode to aggregate the overdue rate acceleration of each reference account age stage to obtain the overdue rate estimated acceleration of the target account age stage;

S32, determining the product of the overdue rate of the appointed account age and the overdue rate estimated acceleration of the target account age as the target overdue rate.

For the account age phases except the first account age phase and the last account age phase in the plurality of account age phases, which are the middle account age phases, the loan risk is still exposed, but the risk release process is stable, and the overdue rate of the current period and the overdue rate of the next period show strong negative correlation, which can be explained as follows: the risk has been fully exposed in the early stages and slowly converged in the later stages. For this, the overdue rate acceleration of the current period can be predicted by referring to the overdue rate acceleration of the same period, that is, the overdue rate acceleration of each reference account age is aggregated by adopting a specified aggregation mode, so as to obtain the overdue rate estimated acceleration of the target account age. In this case, the overdue rate parameter of each reference account age is a rate of increase in the overdue rate of each reference account age. After obtaining the estimated acceleration of the expiration rate of the target account age, the target expiration rate may be determined in a similar manner as described above, which is not described herein.

Optionally, the specified aggregation manner may include a set of aggregation manners (e.g., average acceleration, maximum acceleration, minimum acceleration, etc.), and correspondingly, performing aggregation processing on the expiration rate acceleration of each reference account age by using the specified aggregation manner, where obtaining the expiration rate estimated acceleration of the target account age includes: and respectively carrying out aggregation treatment on the overdue rate acceleration of each reference account age by adopting each aggregation mode in a group of aggregation modes to obtain the overdue rate estimated acceleration corresponding to each aggregation mode, and determining the average value of the overdue rate estimated acceleration corresponding to each aggregation mode as the overdue rate estimated acceleration of the target account age.

Here, as the aggregation methods of multiple calibers such as average acceleration, maximum acceleration, minimum acceleration and the like are adopted, multiple results can be calculated for the same account age, the random forest thought is used for reference, the output values of different calibers are used for fitting, the final result is obtained, the confidence coefficient is improved, and the model is more stable.

For the middle account age stage (for example, the second stage, stage ii), the second prediction model (for example, stage ii model) may be used to predict the overdue rate of the target account age according to the overdue rate acceleration of each reference account age, so as to obtain the target overdue rate, that is, the overdue rate acceleration of each reference account age is input into the second prediction model, so as to obtain the target overdue rate output by the second prediction model. The second prediction model is a prediction model corresponding to other accounting-period phases except for the first accounting-period phase and the last accounting-period phase.

For example, when the account age curve is in the II stage, the data is represented by moderate average value of contract overdue rate, stable average variance of acceleration, and the nth overdue contract overdue rate odr of each month of loan can be used _m,n N+1-term contract expiration rate acceleration inr _m,n+1 Is a result of the history of the same expiration rate odr _m,n And (5) sequencing.

Secondly, dividing the contract overdue rate interval inv according to the percentile _n,j,k ＝(j,k]Where j is the lower limit of the contract overdue interval and k is the upper limit of the contract overdue interval.

Finally, different types of acceleration rates are counted according to different aggregation methodsThe usual aggregation functions have an average acceleration +.>Maximum speed increase +.>Minimum speed-up->Etc. Here, any of the above aggregation statistics may be selectedThe rate of expiration increases.

In determining contract overdue rate, the speed increasing inr 'is estimated' _m,n+1 At that time, the expiration rate odr of the nth period contract of the loan can be put according to the n month _m,n The interval inv where _n,j,k ＝(j,k]Determining the estimated acceleration of the n+1 phase, as shown in a formula (2):

according to the embodiment, in the middle of the account age curve, the corresponding overdue rate acceleration of the account age is used for aggregation treatment to serve as the overdue rate acceleration of the current period, so that the corresponding overdue rate is predicted, and the overdue prediction accuracy can be improved.

In an exemplary embodiment, the above method further comprises:

s41, determining the ratio of the overdue rate of each reference account age to the overdue rate of the last account age of each reference account age as the overdue rate acceleration of each reference account age.

In this embodiment, for the account age in the last account age stage or the account ages in other account ages except the first account age stage and the last account age stage, the ratio between the expiration rate of each reference account age and the expiration rate of the last account age of each reference account age may be determined as the expiration rate acceleration of each reference account age.

For example, for the phase II, the data shows that the average value of the contract overdue rate is moderate, the mean variance of the acceleration is stable, and the contract overdue rate acceleration at the n+1st phase and the contract overdue rate at the n-th phase are in a negative correlation (as shown in FIG. 4), i.e. the higher the contract overdue rate, the lower the contract overdue rate acceleration. Accordingly, to simplify the model, reasonable assumptions can be made using a concept like a first order Markov chain (Markov chain), i.e., assuming a contract expiration rate acceleration inr for the n+1th phase of the ledger function _n+1 Contract expiration Rate odr dependent on the nth phase only _n As shown in formula (2):

inr _m,n+1 ＝f _n+1 (odr _m,n ) (3)

therein, odr _m,n Represents the overdue rate of the nth-period contract of the m-month put loan, inr _m,n Indicating the overdue rate acceleration of the nth period contract of putting loan in m months, f _n (. Cndot.) represents the functional mapping of the rate increase of the n-th contract expiration to the rate of the previous contract expiration.

Here, since the rate of expiration acceleration of the next period is determined depending on the rate of expiration of the last period, even if the rate of expiration of the previous period is overestimated, the previous period measurement deviation is corrected stepwise due to the measurement mechanism of "high rate of expiration matches low rate of expiration", so that a more reasonable result is calculated adaptively.

According to the embodiment, the overdue rate acceleration of the next overdue is related to the overdue rate of the previous overdue, and the overdue rate acceleration of the next overdue is in a negative correlation relationship with the overdue rate acceleration of the next overdue, so that the influence of overdue rate overestimation of the previous overdue can be avoided on the accuracy of the subsequent overdue rate prediction, and the accuracy of overdue rate prediction is improved.

In one exemplary embodiment, determining a target account age stage to which a specified account age of a target credit service belongs includes:

s51, determining the overdue rate acceleration of the appointed account age according to the overdue rate of the last account age of the appointed account age and the overdue rate of the appointed account age;

s52, determining a target account age stage to which the appointed account age belongs according to the overdue rate acceleration of the appointed account age.

In this embodiment, the rate of rise of the specified account age may be determined according to the rate of rise of the last account age of the specified account age and the rate of rise of the specified account age, that is, the ratio of the rate of rise of the specified account age to the rate of rise of the last account age of the specified account age is determined as the rate of rise of the specified account age.

The N account ages of the credit resource are divided into a plurality of account age phases, and for each of the account age phases, there may be a corresponding overdue rate acceleration threshold, and according to the overdue rate acceleration of the designated account age, it may be determined whether the designated account age belongs to each of the account age phases, i.e., the target account age phase to which the designated account age belongs.

According to the embodiment, the account age stage of each account age is determined according to the overdue rate acceleration of each account age, so that overdue rate prediction is performed on the account ages of different account age stages by adopting different overdue rate prediction modes, and the accuracy of overdue rate prediction can be improved.

In one exemplary embodiment, determining the target account age stage to which the specified account age belongs according to the overdue rate acceleration of the specified account age includes:

s61, determining a target account age stage to which the appointed account age belongs as a first account age stage under the condition that the overdue rate acceleration of the appointed account age is greater than or equal to a first acceleration threshold;

s62, determining a target account age stage to which the appointed account age belongs as a second account age stage under the condition that the overdue rate acceleration of the appointed account age is larger than or equal to a second acceleration threshold and smaller than a first acceleration threshold;

s63, determining a target account age stage to which the appointed account age belongs as a third account age stage under the condition that the overdue rate acceleration of the appointed account age is smaller than a second acceleration threshold;

the first speed increasing threshold value is larger than the second speed increasing threshold value, and the N account age periods are sequentially divided into a first account age period, a second account age period and a third account age period according to the sequence from front to back.

The account age stage of the credit business can be divided into a first account age stage, a second account age stage and a third account age stage from front to back, and each account age stage corresponds to a different overdue rate acceleration threshold. The N account ages may be divided into a first account age phase, a second account age phase, and a third account age phase in order from front to back, where the first account age phase, the second account age phase, and the third account age phase may correspond to an initial account age increase period, an intermediate account age increase period, and an end account age increase period, respectively.

Under the condition that the overdue rate acceleration of the appointed account age is larger than or equal to a first acceleration threshold value, determining a target account age stage to which the appointed account age belongs as a first account age stage; under the condition that the overdue rate acceleration of the appointed account age is larger than or equal to a second acceleration threshold value and smaller than a first acceleration threshold value, the target account age stage to which the appointed account age belongs can be determined to be a second account age stage; under the condition that the overdue rate acceleration of the appointed account age is smaller than the second acceleration threshold, the target account age stage to which the appointed account age belongs can be determined to be the third account age stage, wherein the first acceleration threshold and the second acceleration threshold can be determined by clustering the overdue rate mean value and the ring ratio acceleration standard deviation of a plurality of credit businesses in each account age, and can be set manually, and the method is not limited in the embodiment.

The first account age stage is an initial account age growth stage, the overdue rate acceleration of each account age of the first account age stage is relatively high, the second account age stage is an intermediate account age growth stage, and the overdue rate acceleration of each account age of the second account age stage is reduced relative to the overdue rate acceleration of each account age of the first account age stage, so that the first credit threshold is greater than the second acceleration threshold.

Optionally, the first accounting period (early accounting period of growth), the second accounting period (middle accounting period) and the third accounting period (end accounting period) are similar to the first, second and third accounting periods in the foregoing embodiments, and are not described here.

According to the embodiment, according to the overdue rate acceleration of each account age and the threshold value of each credit stage, each account age of the credit business is divided into different account age stages, so that the overdue rate of the account age of the different account age stages is predicted, and the accuracy of business information prediction can be improved.

In an exemplary embodiment, the above method further comprises:

s71, clustering the average value of the overdue rates and the ring ratio acceleration standard deviation of N account age periods of the target credit resource issued in a group of historical time periods, and dividing the N account age periods of the target credit resource issued in a group of historical time periods into account age class clusters corresponding to each of a plurality of account age periods;

s72, determining a threshold value of the overdue rate acceleration corresponding to each account age stage according to the overdue rate acceleration of the account ages in the account age class cluster corresponding to each account age stage.

In this embodiment, the overdue rate of the target credit resource issued in the set of history periods in the N account ages may be obtained, where the history period may be a period before the issue of the target credit resource corresponding to the target credit service, and the target credit resource issued in the set of history periods is the target credit resource issued before the issue of the target credit resource corresponding to the target credit service. According to the overdue rates of the target credit resources issued in the historical time periods in the N account age periods, the average value of the overdue rates and the ring ratio acceleration standard deviation of the N account age periods of the target credit resources issued in the historical time periods can be obtained.

The average value of the overdue rates and the standard deviation of the ring ratio of the N account age periods of the target credit resource issued in a group of historical time periods are clustered, the average value of the overdue rates of the N account age periods of the target credit resource issued in a group of historical time periods are similar to the account age periods of the ring ratio of the N account age periods of the target credit resource issued in a group of historical time periods, the N account age periods of the target credit resource issued in a group of historical time periods can be divided into account age class clusters corresponding to each of a plurality of account age stages, wherein the average value of the overdue rates of the account age periods in the account age class clusters corresponding to each of the plurality of account age stages are similar, and the standard deviation of the ring ratio of the account age periods is similar.

According to the overdue rate acceleration of the account age in the account age class cluster corresponding to each account age stage, the overdue rate acceleration threshold corresponding to each account age stage, that is, the first acceleration threshold and the second acceleration threshold, may be determined.

For example, an account age table of a plurality of credit businesses is obtained, the overdue rate acceleration of each credit business at each account age is calculated, the contract overdue rate (i.e., overdue rate) of each credit business at each account age and the mean value and standard deviation corresponding to the overdue rate acceleration are calculated, the overdue rate mean value and the ring ratio acceleration standard deviation of the plurality of credit businesses at each account age are clustered to determine the overdue rate acceleration threshold of each growth stage (i.e., account age stage), or the overdue rate acceleration threshold of each growth stage is manually set to determine the growth stage to which the specified account age belongs.

According to the embodiment, the threshold value of the accelerated overdue rate of each account age stage is determined according to the overdue rates of a plurality of credit businesses in different account ages, so that the account age stage to which each account age belongs is determined, and the accuracy of overdue rate prediction can be improved.

Illustratively, taking dividing the account age curve into three stages as an example, an algorithm for efficiently predicting and deducing the account age Vintage function of the credit product based on the divide-and-conquer chain concept can be adopted, and referring to fig. 6, the main processing flow of the overdue rate prediction includes: for the processing flow, training data are input in the model training process, the overdue rate acceleration and descriptive statistical indexes thereof are calculated, then the growth stage is divided, independent modeling is carried out according to models of different stages, model parameters are stored after completion, and account age curves are predicted, so that the problems of low efficiency, more manual intervention, poor generalization capability and the like of the overdue rate prediction method in the related technology are solved.

For data preprocessing, the training sample of the model is an account age table, the data form of the training sample can be a two-dimensional table, the row of the account age table can represent loans issued in a certain period (such as month by month and year by year), and the column of the account age table can represent account age numbers from 1 st period to loan deadline. The cell at (m, n) of the table of ages means: and a loan is issued in the m period, and the contract overdue rate at the nth period.

After the account age table is input, thresholds corresponding to different account age phases are determined based on information recorded in the account age table to determine an increasing phase in which the term account age function is located. In this example, the growth of the account age curve is divided into 3 phases, i.e., phases i, ii, and iii. The model is modeled according to different strategies for different stages.

Here, the model uses divide-and-conquer to model the different stages of the account age curve separately. In order to objectively divide the growth stage of the account age curve, the variance of the account age overdue rate acceleration can be used for determining, in an ideal state, the account age Vintage function is a convex function, the initial growth is fast, the risk is gradually exposed along with the growth of the account age, and the account age final period gradually converges, so that the variance of the account age acceleration is expressed to be fast and slow firstly and finally tends to 0, and therefore, the variance can be used for stage division.

For the model construction, the stage I model construction, the stage II model construction and the stage III model construction can be respectively carried out, wherein when the stage II model construction is carried out, the steps of generating an interval acceleration reference table, determining the estimated acceleration of the contract overdue rate and calculating the estimated contract overdue rate of the n+1th period according to a formula can be repeatedly executed, and the contract overdue rate of each period is calculated in a recurrence manner.

By the optional example, the problems of low efficiency, more manual intervention, poor generalization capability and the like of the overdue rate prediction mode in the related technology can be solved, and the prediction efficiency and the generalization capability of the model are improved.

It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required for the present application.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM (Read-Only Memory)/RAM (Random Access Memory), magnetic disk, optical disk), comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present application.

According to another aspect of the embodiment of the present application, there is also provided a service information prediction apparatus for implementing the above service information prediction method. Fig. 7 is a block diagram illustrating a structure of an alternative service information prediction apparatus according to an embodiment of the present application, and as shown in fig. 7, the apparatus may include:

a first determining unit 702, configured to determine a target account age stage to which a designated account age of a target credit service belongs, where the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N;

a second determining unit 704, connected to the first determining unit 702, configured to determine a timeout rate parameter of each reference account age in a set of reference account ages matched with the target account age, where the target account age is a next account age of the designated account age, and the set of reference account ages includes (n+1) th account ages of the target credit resource issued in the 1 st to (m-1) th periods;

the prediction unit 706 is connected to the second determination unit 704, and is configured to predict the overdue rate of the target account age according to a target prediction mode according to the overdue rate parameter of each reference account age, so as to obtain the target overdue rate, where the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

It should be noted that, in this embodiment, the first determining unit 702 may be used to perform the step S202, the second determining unit 704 in this embodiment may be used to perform the step S204, and the predicting unit 706 in this embodiment may be used to perform the step S206.

Through the module, determining a target account age stage to which a designated account age of a target credit service belongs, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N; determining overdue rate parameters of each reference account age in a group of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the group of reference account ages comprises (n+1) th account ages of target credit resources issued from a 1 st time period to a (m-1) th time period; according to the overdue rate parameter of each reference account age, the overdue rate of the target account age is predicted according to a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is the overdue rate prediction mode corresponding to the target account age stage, different account age stages correspond to different overdue rate prediction modes, the problem that the service information prediction accuracy is low in the service information prediction method in the related technology is solved, and the service information prediction accuracy is improved.

In one exemplary embodiment, the prediction unit includes:

the first determining module is configured to determine, as the target overdue rate, a mean value of overdue rates corresponding to each reference account age in a case where the target account age stage is a first account age stage of the plurality of account age stages.

In one exemplary embodiment, the prediction unit includes:

the second determining module is used for determining the average value of the overdue rate acceleration of each reference account age to obtain the average value of the overdue rate acceleration, and determining the standard value of the overdue rate acceleration of each reference account age to obtain the standard value of the overdue rate acceleration when the target account age stage is the last account age stage of the plurality of account age stages;

the third determining module is used for determining a target normal distribution function according to the overdue rate acceleration average value and the overdue rate acceleration standard value, wherein the overdue rate acceleration average value is a position parameter of the target normal distribution function, the overdue rate acceleration standard value is a shape parameter of the target normal distribution function, and the target normal distribution function is a normal distribution function between the overdue rate and the overdue rate acceleration;

the fourth determining module is used for determining a function value corresponding to a random expected rate in a target normal distribution function and overdue rate range as overdue rate estimated acceleration of a target account age, wherein the overdue rate range is the overdue rate of each reference account age;

And the fifth determining module is used for determining the product of the overdue rate of the appointed account age and the overdue rate estimated acceleration of the target account age as the target overdue rate.

In one exemplary embodiment, the prediction unit includes:

the aggregation module is used for carrying out aggregation processing on the overdue rate acceleration of each reference account age by adopting a designated aggregation mode under the condition that the target account age stage is an account age stage except for a first account age stage and a last account age stage in a plurality of account age stages, so as to obtain the overdue rate estimated acceleration of the target account age;

and the sixth determining module is used for determining the product of the overdue rate of the appointed account age and the overdue rate estimated acceleration of the target account age as the target overdue rate.

In an exemplary embodiment, the above apparatus further includes:

and the third determining unit is used for determining the ratio between the overdue rate of each reference account age and the overdue rate of the last account age of each reference account age as the overdue rate acceleration of each reference account age.

In one exemplary embodiment, the first determining unit includes:

the seventh determining module is used for determining the overdue rate acceleration of the appointed account age according to the overdue rate of the last account age of the appointed account age and the overdue rate of the appointed account age;

And the eighth determining module is used for determining a target account age stage to which the appointed account age belongs according to the overdue rate acceleration of the appointed account age.

In one exemplary embodiment, the eighth determination module includes:

the first determining submodule is used for determining that a target account age stage to which the appointed account age belongs is a first account age stage under the condition that the overdue rate acceleration of the appointed account age is larger than or equal to a first acceleration threshold value;

the second determining submodule is used for determining that the target account age stage to which the appointed account age belongs is a second account age stage under the condition that the overdue rate acceleration of the appointed account age is larger than or equal to a second acceleration threshold value and smaller than the first acceleration threshold value;

the third determining submodule is used for determining that the target account age stage to which the appointed account age belongs is a third account age stage under the condition that the overdue rate acceleration of the appointed account age is smaller than a second acceleration threshold value;

In an exemplary embodiment, the above apparatus further includes:

the dividing unit is used for dividing the N account age groups of the target credit resource issued in the historical time period into account age class clusters corresponding to each of a plurality of account age stages by clustering the average value of the overdue rates and the ring ratio acceleration standard deviation of the N account age groups of the target credit resource issued in the historical time period;

And the fourth determining unit is used for determining the overdue rate acceleration threshold corresponding to each account age stage according to the overdue rate acceleration of the account ages in the account age class cluster corresponding to each account age stage.

It should be noted that the above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments. It should be noted that the above modules may be implemented in software or in hardware as part of the apparatus shown in fig. 1, where the hardware environment includes a network environment.

According to yet another aspect of an embodiment of the present application, there is also provided a storage medium. Alternatively, in this embodiment, the storage medium may be used to execute the program code of the prediction method of the service information of any of the above embodiments of the present application.

Alternatively, in this embodiment, the storage medium may be located on at least one network device of the plurality of network devices in the network shown in the above embodiment.

Alternatively, in the present embodiment, the storage medium is configured to store program code for performing the steps of:

s1, determining a target account age stage to which a designated account age of a target credit service belongs, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N;

S2, determining overdue rate parameters of each reference account age in a group of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the group of reference account ages comprises the (n+1) th account age of a target credit resource issued from the 1 st time period to the (m-1) th time period;

s3, predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

Alternatively, specific examples in the present embodiment may refer to examples described in the above embodiments, which are not described in detail in the present embodiment.

Alternatively, in the present embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a U disk, ROM, RAM, a mobile hard disk, a magnetic disk or an optical disk.

According to still another aspect of the embodiments of the present application, there is also provided an electronic device for implementing the above-mentioned service information prediction method, where the electronic device may be a server, a terminal, or a combination thereof.

Fig. 8 is a block diagram of an alternative electronic device, according to an embodiment of the present application, as shown in fig. 8, including a processor 802, a communication interface 804, a memory 806, and a communication bus 808, wherein the processor 802, the communication interface 804, and the memory 806 communicate with each other via the communication bus 808, wherein,

a memory 806 for storing a computer program;

the processor 802, when executing the computer program stored on the memory 806, performs the following steps:

Alternatively, the communication bus may be a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or an EISA (Extended Industry Standard Architecture ) bus, or the like. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, only one thick line is shown in fig. 8, but not only one bus or one type of bus. The communication interface is used for communication between the electronic device and other equipment.

The memory may include RAM or may include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.

As an example, the above memory 806 may be, but not limited to, a first determining unit 702, a second determining unit 704, and a predicting unit 706 in a predicting apparatus including the above traffic information. In addition, other module units in the prediction apparatus of the service information may be included, but are not limited to, and are not described in detail in this example.

The processor may be a general purpose processor and may include, but is not limited to: CPU (Central Processing Unit ), NP (Network Processor, network processor), etc.; but also DSP (Digital Signal Processing, digital signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field-Programmable Gate Array, field programmable gate array) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.

It will be understood by those skilled in the art that the structure shown in fig. 8 is only schematic, and the device implementing the method for predicting service information may be a terminal device, and the terminal device may be a smart phone (such as an Android mobile phone, an iOS mobile phone, etc.), a tablet computer, a palmtop computer, a mobile internet device (Mobile Internet Devices, MID), a PAD, etc. Fig. 8 is not limited to the structure of the electronic device. For example, the electronic device may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 8, or have a different configuration than shown in FIG. 8.

Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program for instructing a terminal device to execute in association with hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: flash disk, ROM, RAM, magnetic or optical disk, etc.

The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

The integrated units in the above embodiments may be stored in the above-described computer-readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing one or more computer devices (which may be personal computers, servers or network devices, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.

In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In several embodiments provided by the present application, it should be understood that the disclosed client may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in the present embodiment.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or at least two units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application, which are intended to be comprehended within the scope of the present application.

Claims

1. A method for predicting service information, comprising:

determining a target account age stage to which a designated account age of a target credit service belongs, wherein the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N;

determining overdue rate parameters of each reference account age in a set of reference account ages matched with a target account age, wherein the target account age is the next account age of a designated account age, and the set of reference account ages comprises (n+1) th account ages of the target credit resource issued from a 1 st time period to a (m-1) th time period; and predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

2. The method according to claim 1, wherein predicting the expiration rate of the target account age according to the expiration rate parameter of each reference account age in a target prediction manner to obtain a target expiration rate comprises:

and determining a mean value of the overdue rate corresponding to each reference account age as the target overdue rate when the target account age stage is a first account age stage of the plurality of account age stages.

3. The method according to claim 1, wherein predicting the expiration rate of the target account age according to the expiration rate parameter of each reference account age in a target prediction manner to obtain a target expiration rate comprises:

under the condition that the target account age stage is the last account age stage in the plurality of account age stages, determining the average value of the overdue rate acceleration of each reference account age stage to obtain the average value of the overdue rate acceleration, and determining the standard value of the overdue rate acceleration of each reference account age stage to obtain the standard value of the overdue rate acceleration; determining a target normal distribution function according to the overdue rate acceleration average value and the overdue rate acceleration standard value, wherein the overdue rate acceleration average value is a position parameter of the target normal distribution function, the overdue rate acceleration standard value is a shape parameter of the target normal distribution function, and the target normal distribution function is a normal distribution function between overdue rate and overdue rate acceleration;

Determining a function value corresponding to a random expected rate in the target normal distribution function and the overdue rate range as the overdue rate estimated acceleration of the target account age, wherein the overdue rate range comprises the overdue rate of each reference account age;

and determining the product of the overdue rate of the designated account age and the estimated acceleration of the overdue rate of the target account age as the target overdue rate.

4. The method according to claim 1, wherein predicting the expiration rate of the target account age according to the expiration rate parameter of each reference account age in a target prediction manner to obtain a target expiration rate comprises:

under the condition that the target account age stage is an account age stage except a first account age stage and a last account age stage in the plurality of account age stages, adopting a designated aggregation mode to aggregate the overdue rate acceleration of each reference account age stage to obtain the overdue rate estimated acceleration of the target account age stage;

5. The method according to claim 3 or 4, characterized in that the method further comprises:

And determining the ratio of the overdue rate of each reference account age to the overdue rate of the last account age of each reference account age as the overdue rate acceleration of each reference account age.

6. The method of any one of claims 1 to 4, wherein the determining the target account age stage to which the specified account age of the target credit service belongs comprises:

determining the overdue rate acceleration of the appointed account age according to the overdue rate of the last account age of the appointed account age and the overdue rate of the appointed account age;

and determining the target account age stage to which the appointed account age belongs according to the overdue rate acceleration of the appointed account age.

7. The method of claim 6, wherein the determining the target account age stage to which the specified account age belongs according to the rate acceleration of the overdue rate of the specified account age comprises:

determining the target account age stage to which the designated account age belongs as a first account age stage under the condition that the overdue rate acceleration of the designated account age is greater than or equal to a first acceleration threshold;

determining that the target account age stage to which the designated account age belongs is a second account age stage when the overdue rate acceleration of the designated account age is greater than or equal to a second acceleration threshold and less than the first acceleration threshold;

Determining that the target account age stage to which the designated account age belongs is a third account age stage under the condition that the overdue rate acceleration of the designated account age is smaller than a second acceleration threshold;

the first speed increasing threshold is larger than the second speed increasing threshold, and the N account ages are sequentially divided into the first account age stage, the second account age stage and the third account age stage according to the sequence from front to back.

8. The method of claim 7, wherein the method further comprises:

clustering the average value of the overdue rates and the ring ratio acceleration standard deviation of N account age of the target credit resource issued in a group of history time periods, and dividing the N account age of the target credit resource issued in the group of history time periods into account age class clusters corresponding to each of the plurality of account age stages;

and determining a threshold value of the overdue rate acceleration corresponding to each account age stage according to the overdue rate acceleration of the account ages in the account age class cluster corresponding to each account age stage.

9. A device for detecting service information, comprising:

a first determining unit, configured to determine a target account age stage to which a designated account age of a target credit service belongs, where the designated account age is an nth account age of a target credit resource corresponding to the target credit service issued in an mth period, N account ages of the target credit resource are divided into a plurality of account age stages, m and N are positive integers greater than or equal to 1, and N is a positive integer greater than or equal to N;

A second determining unit, configured to determine an overdue rate parameter of each reference account age in a set of reference account ages matched with a target account age, where the target account age is a next account age of a designated account age, and the set of reference account ages includes (n+1) th account age of the target credit resource issued in a 1 st to (m-1) th period;

the prediction unit is used for predicting the overdue rate of the target account age according to the overdue rate parameter of each reference account age in a target prediction mode to obtain the target overdue rate, wherein the target prediction mode is a overdue rate prediction mode corresponding to the target account age stage, and different account age stages correspond to different overdue rate prediction modes.

10. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 8.

11. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to perform the method of any of claims 1 to 8 by means of the computer program.