KR20160036515A - System and Method for Real time based Credit Risk Analysis using the Integrated Account Information - Google Patents

Abstract

Disclosed is a computer-readable recording medium recording a program to execute a method to analyze credit risk in real time based on integrated account information. The method includes: a step of classifying registration accounts by type in a computer; a step of generating integrated account information about the registration accounts, classified by type; a step of calculating multiple credit risk indexes about an integrated account based on the integrated account information; and a step of calculating a comprehensive credit score from the credit risk indexes.

Description

SYSTEM AND METHOD FOR REAL-TIME CREDIT RISK ANALYSIS USING THE INTEGRATED ACCOUNT INFORMATION BASED ON CRITERIA

The present invention relates to a real-time credit risk analysis method and system based on integrated account information.

Credit risk is the probability that a creditor will suffer losses due to the default of a financial lender transaction customer or debtor, such as an individual or a corporation. In the case of a financial institution, It is used to calculate the appropriate loan rate considering the decision on approval and risk.

Since these risks vary according to the type of customer and the type of transaction or the terms of the transaction, the measurement model and management standard should be applied separately in each case. For example, in the case of individual clients, the size and nature of loans are relatively homogeneous and small, while corporate clients have large transactions, so most lending is based on collateral rather than pure credit. The average loss rate of credit risk However, the risk-loss distribution is high, so subjective judgment of individual customers is relatively important.

Applying these principles, the current credit risk system is largely based on financial information-based models (eg interest compensation costs, Altman's ZETA model, Logit model, etc.) or financial engineering theory model (eg KMV model using stock information (Eg, regression analysis, logistic regression, neural network, etc.) based on a credit risk management system and a statistical method (eg, JPMorgan model based on transition matrix) (Credit Scoring System) that predicts individual default probability of default.

However, these current credit risk systems have several problems.

First, the timeliness of evaluation is low and only limited application is possible for SMEs.

In the corporate credit risk system, semi-annual financial information, non-financial information, and market information such as stock prices are used as input variables. Because these input variables are different from each other, even if credit risk is measured at present, Information, but most of the remaining variables use historical information, making fair assessment of the present time difficult. In addition, the credibility of the evaluation is very low because the credibility is low even if there is no certified financial information, which is a necessary consideration item in the case of SMEs.

Second, only the debt information excluding the asset items of customers' financial information is evaluated.

The personal credit rating system uses only the information related to the receipt of personal financial transaction information, that is, excluding the asset item and the credit, that is, debt. However, considering that credit risk is the probability of loss due to default, it is necessary to consider the customer's asset information, which is the standard of the debt repayment ability, as an input variable in the evaluation model in order to ascertain the probability of default. The system has a limitation that it considers only debt information and delinquency information.

Third, new trading customers are limited to limited information.

The personal credit rating system is classified into the Application Scoring Model (ASS) applied at the time of applying for the loan and the Behavioral Scoring Model (BSS) using the past performance information of the transaction customer after handling. Most of them use only at the level of the customer database, and actual financial institutions place a greater weight on the BSS. However, since new transaction customers have little accumulated information internally in financial institutions, it is possible to apply the BSS from a point in time after the transaction has occurred.

According to an embodiment of the present invention, by generating new real-time-based information called integrated account information using all financial assets and debt information of a customer's name at home and abroad, A real-time credit risk analysis method and system capable of performing proactive management based on the real-time credit risk, and a computer-readable recording medium on which a program for executing the real-time credit risk analysis method and system can be provided.

According to an embodiment of the present invention, the integrated account information is generated using all the reception information of customers scattered in domestic and foreign financial institutions. Therefore, , It is possible to manage the credit risk from the time of the first transaction. Furthermore, it is possible to grasp the transfer phenomenon of the credit risk among individual financial institutions and the transition phenomenon between financial product types in real time Therefore, a real-time credit risk analysis method, system, and program for implementing such a system can reduce the risk to be borne by the financial institution itself by ultimately recognizing the change history of the customer's credit risk and performing consulting or management accordingly A computer readable recording medium having recorded thereon can be provided.

The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood from the following description.

According to an embodiment of the present invention,

On the computer,

Classifying the registered accounts by type;

Generating consolidated account information for registered accounts classified by type;

Calculating a plurality of credit risk indices for the unified account based on the unified account information; And

And a step of calculating a total credit score from the credit risk indices based on the credit risk indexes. The computer-readable recording medium records a program for executing the real-time credit risk analysis method based on the integrated account information.

According to another embodiment of the present invention,

In a real-time credit risk analysis system based on integrated account information,

A server of a financial institution that stores information about registered accounts; And

A plurality of credit risk indices are calculated for the unified account based on the unified account information, and the credit risk indices are calculated based on the unified account information, And a credit risk analysis server for calculating a total credit score from the credit risk analysis server.

According to an embodiment of the present invention, the credit risk indexes may include at least one of a safety factor, a default probability, and a deposit turnover rate.

According to one embodiment of the present invention, the types of registration accounts may be comprised of demand deposits, savings deposits, loans, funds, insurance, stocks, stocks, and bonds.

According to an embodiment of the present invention,

The step of calculating the credit risk indices comprises:

Calculating a deposit turnover rate for each type of registered account,

The turnover rate may be defined as a value obtained by dividing the total withdrawal amount of an account for a unit time by the average over the same period.

According to an embodiment of the present invention,

The safety factor,

, Where m is the safety factor, x is the withdrawal amount of the consolidated account, y is the cumulative amount of the unit time of the consolidated account,

를 만족하는 정규분포확률변수이고,

는 x에 대한 평균이고,

는 x 에 대한 표준편차이며,x is

Is a normal distribution random variable satisfying the following equation

Is the mean for x,

Is the standard deviation for x,

를 만족하는 정규분포확률변수이고,

는 y에 대한 평균이고,

는 y에 대한 표준편차일 수 있다. y is

Is a normal distribution random variable satisfying the following equation

Is the mean for y,

May be the standard deviation for y.

According to an embodiment of the present invention,

The default probability

Lt; / RTI >

,

,

,

,

,

,

Lt; / RTI >

According to one or more embodiments of the present invention, by generating new information in real-time based on integrated account information using all financial assets and debt information of a customer's name at home and abroad, Real-time based proactive management can be performed.

According to one or more embodiments of the present invention, the integrated account information is generated by using all the reception information of customers scattered in domestic and foreign financial institutions. Therefore, In addition, it is possible to manage the credit risk from the point of the first transaction because the credit information of the other financial institutions can be used for the customer. Furthermore, the credit risk transfer phenomenon between the individual financial institutions and the transition phenomenon between the financial product types can be grasped in real time Therefore, it is effective to reduce the risk to be borne by the financial institution itself by ultimately recognizing the change history of the customer's credit risk and performing consulting or management accordingly.

1 is a diagram for explaining a real-time credit risk analysis system based on integrated account information according to an embodiment of the present invention.
2 is a diagram illustrating a configuration of a real-time credit risk analysis server according to an embodiment of the present invention.
3 shows an example of a code definition for an account number system of a financial institution.
4 is a flowchart illustrating a credit risk management method based on integrated account information according to an embodiment of the present invention.
FIG. 5 is a diagram for explaining a method for generating consolidated account information by type according to an embodiment of the present invention.
FIG. 6 is a diagram for explaining a method of generating a default probability and a safety factor according to an embodiment of the present invention. Referring to FIG.
FIG. 7 is a view for explaining a method of generating a deposit turnover rate according to an embodiment of the present invention.
8 is a diagram for explaining a method of generating a total credit score according to an embodiment of the present invention.
FIG. 9 is a diagram showing the classification of financial products by type (Table 1) and the termination / redemption order of financial products (Table 2).
10 is a diagram showing an example (Table 3) of generation of a consolidated account number for each product type.
FIG. 11 is a diagram showing the relationship between the default, the amount of withdrawal, and the number of redemptions using the load-intensity model (Table 4).
12 is a diagram illustrating mathematical formulas for explaining embodiments of the present invention.
FIG. 13 is a diagram showing the relationship between the withdrawal order and the turnover rate of the remittance account.
14 is a graph showing an example of credit risk scaling criteria using the deposit turnover rate and the beta (Table 6).
FIG. 15 is a view showing the definition of a default event using probability distribution characteristics of a withdrawal amount and an appropriate number.
16 is a graph showing an example of credit risk scaling criteria for the default probability and the safety factor (Table 8).

BRIEF DESCRIPTION OF THE DRAWINGS The above and other objects, features, and advantages of the present invention will become more readily apparent from the following description of preferred embodiments with reference to the accompanying drawings. However, the present invention is not limited to the embodiments described herein but may be embodied in other forms. Rather, the embodiments disclosed herein are provided so that the disclosure can be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

The terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. The terms "comprises" and / or "comprising" used in the specification do not exclude the presence or addition of one or more other elements.

Hereinafter, the present invention will be described in detail with reference to the drawings. In describing the specific embodiments below, various specific details have been set forth in order to explain the invention in greater detail and to assist in understanding it. However, it will be appreciated by those skilled in the art that the present invention may be understood by those skilled in the art without departing from such specific details. In some instances, it should be noted that portions of the invention that are not commonly known in the description of the invention and are not significantly related to the invention do not describe confusing reasons to explain the present invention.

For the embodiment of the present invention, some terms will be defined as follows. First, the scope of the financial transaction customer is limited to an individual or a small and medium-sized enterprise except for a large company in this embodiment, and a financial product type covers all the credit and receiving accounts of a bank.

Credit risk is the probability that a creditor will suffer losses due to default by a debtor. In particular, the reason why the measurement and management of credit risk is necessary in a financial institution is as follows. First, (Default), it is necessary to calculate the minimum equity capital that normal business activity can be done. Secondly, it is necessary to decide whether to approve the financial loan transaction such as loan application and calculate the appropriate loan rate considering the risk Because it is necessary.

Since these risks vary according to the type of customer and the type of transaction or the terms of the transaction, the measurement model and management standard should be applied separately in each case.

Although there have been many quantitative researches on the credit risk related to credit institutions of financial institutions, most of them are centered on large corporate loans, and studies on small-scale individual or SME loans are very poor, Is a level in which some evaluation models suitable for large companies are used. The reason for this is that in terms of the total risk of the financial institution, the portion of the corporate loan is larger than that of an individual or SME loan, and the number of the corporate loan is smaller than that of an individual or small business loan. While credit risk is difficult to measure because the number of individuals or SME loans is too large and the size of the loan is small.

In recent years, however, financial institutions focused on corporate finance have been expanding the retail banking of individuals or SMEs in order to expand their profitability. At the same time, the government's policy to support retail banking, Retail financing is becoming more active due to demand side effects such as rapid changes in lifestyle and values. As a result of the sudden expansion of retail financing, excessive debt risk, massive credit delinquency, and insolvency of financial institutions due to the deterioration of retail financing are obstacles to the development of domestic retail finance and securing the competitiveness of financial institutions. Institutions are increasingly in need of establishing a credit rating system suitable for retail loans and operating them effectively.

In the case of individual clients, financial institutions are using the Credit Scoring System (CSS) to manage credit risk based on credit transactions, and SMEs are using corporate credit evaluation models. Among them, the credit rating system is classified into the Application Scoring Model (ASS) applied at the time of applying for the loan and the Behavioral Scoring Model (BSS) , Most of them use only at the level of customer database, and the actual financial institutions place a greater weight on BSS. However, in the case of new trading customers, it is practically impossible to apply BSS because there is little accumulated information about counterparties inside the financial institution.

In addition to these problems, the problems presented so far in the credit risk assessment process for credit transactions of tens of millions of individual and SMEs by financial institutions are summarized as follows.

1) Actual default rate and expected loss are higher than general corporate loans.

2) Due to the large number of accounts, the default correlation is lower than the corporate exposure.

3) Unexpected losses are lower than corporate exposures.

4) Credit risk does not increase sharply because the impact of default on some of the loans is small and the default correlation is low.

5) The frequency of losses is higher in the retail sector, but the loss severity is lower than in the corporate sector.

6) In the case of SMEs, verified financial statements are not prepared like general companies. Even if they are prepared, the cycle is very low, about once a year.

7) SMEs have limited information available for evaluation, such as news, production / operation information, etc., other than financial information, and the on-site inspection is almost impossible.

Due to these various problems, in the embodiment of the present invention, instead of utilizing existing evaluation models and input variables in the credit risk evaluation of an individual or a small and medium-sized enterprise, all the receiving accounts and general financial account accounts of customers scattered in domestic and overseas financial institutions The credit risk information is automatically classified according to the product type defined in the present invention, screen scraping is performed, and the integrated credit information is generated using the collected transaction information to measure real-time credit risk.

The reason for not using the existing evaluation models and input variables is that the portfolios of assets and liabilities are not diversified as in the case of general companies in the case of individuals or SMEs, and most of them are concentrated in the receiving transactions with financial institutions. This is because it can be checked at the earliest through account receipt and balance details.

According to one embodiment of the present invention, when a customer registers an account number of the entire financial product in his / her name instead of an input variable such as financial information used in the existing model, the registered account is automatically classified as a product type defined in the present invention , Screen scraping for each account is performed to collect real-time transaction information for each account, and based on the transaction information, the integrated account information for each product type and the integrated account information for the entire account are automatically generated based on the transaction information, Real-time safety factor, and real-time deposit turnover rate, it is possible to measure the customer's credit risk and credit risk transition in real time.

For this purpose, according to an embodiment of the present invention, after generating integrated account information for each product type or the whole, rather than an individual financial account, by utilizing the concept of the integration account balance and the balance account, And solved the problem of using only the debt information by using the consolidated account information reflecting all of the customer's financial assets and liabilities information, By using information, it is possible to overcome limitations of evaluation for new customers due to lack of information.

FIG. 1 illustrates a real-time credit risk analysis system based on integrated account information according to an embodiment of the present invention. FIG. 2 illustrates a configuration of a real-time credit risk analysis server according to an embodiment of the present invention. to be.

The real-time credit risk analysis system based on the integrated account information according to an embodiment of the present invention is configured to store and manage information on the communication network 10, the credit risk analysis server 100, the member terminal 200, (Deletion, modification, addition) of the financial institution.

In this specification, the term 'financial institution' shall be used to refer to an institution having account information on demand deposits, savings deposits, loans, funds, shares, and bonds.

The communication network 10 refers to various networks that support wired and / or wireless communication.

According to the present embodiment, the credit risk analysis server 100 can perform real-time credit risk analysis operations by acquiring account information from the server 300 of the financial institution.

According to an embodiment of the present invention, the credit risk analysis operation is performed by, for example, classifying the registered accounts by type, generating unified account information for the registered accounts classified by type, Calculating a plurality of credit risk indices for the integrated account, and calculating a total credit score from the credit risk indices.

According to an embodiment of the present invention, the credit risk analysis server 100 includes hardware such as a memory (not shown), a computer processor (not shown), and a storage unit for storing various data, and a control unit, It includes basic programs such as Web page management and software such as programs that can perform credit risk analysis operations in real time.

The storage unit may store a member DB, an account number master DB, a financial product DB, a safety factor scaling DB, a turnover scaling DB, and a bankruptcy probability scaling DB.

The member DB contains the member's personal information.

The account number master DB contains the patterns of the account number. For example, since the patterns of the account numbers may be different for banks, insurance companies, and securities companies, these patterns are included in the account number mast DB. That is, the account number mast DB includes data indicating which items the account numbers of the financial institution are composed.

The financial product DB includes data classified by types of financial products. For example, a financial instrument DB contains data that classifies all financial instruments issued by a financial institution, such as a demand deposit or a savings account of a bank, and a securities account or an insurance account.

The turnover scaling DB includes deposit turnover scaling criteria.

The default probability scaling DB includes a default probability scaling criterion.

A program that performs a real-time credit risk analysis operation may be loaded into a memory under the control of a computer processor, and may use DBs stored in a storage unit.

The program for performing the real-time credit risk analysis operation according to an exemplary embodiment of the present invention includes a membership registration / authentication module, a financial account classification module, a screen scraping module, a unified account information generation module, a safety factor calculation module, A real-time credit risk analysis operation such as a probability calculation module, a statistic calculation module, and a total credit score calculation module.

The membership registration / authentication module is a module for providing a user input / output interface for membership registration and authentication.

The financial account classification module for each type is a module for classifying financial accounts held by members by type. The type-specific financial account classification module can classify the account type by using, for example, 'subject' information included in the account information registered by the user.

The screen scraping module acquires transaction information (including transaction history information of the corresponding account, for example, transaction date, transaction date and time, transaction amount, and transaction purpose) of the financial account registered by the user from the server of the financial institution Module.

The integrated account information generation module can generate integrated account information for each type. In addition, the integrated account information generation module can re-integrate the integrated account information by type to generate the entire integrated account information.

The safety factor calculation module can calculate the safety factor, the turnover rate module can calculate the turnover rate, and the default rate calculation module can calculate the default rate.

The statistical calculation module can calculate statistical values (for example, average or standard deviation) using transaction amounts or balance information for individual accounts or for integrated accounts.

 The total credit score calculation module can calculate the total credit score based on the safety factor, turnover rate, and default probability. For example, a composite credit score can be calculated by assigning and summing individual ratings on credit risk indices.

More specifically, the aggregate credit score can be calculated by adding and summing individual ratings for each of the safety factor, default probability, and deposit turnover rate.

According to one embodiment of the present invention, the credit risk indexes include at least one of a safety factor, a default probability, and a deposit turnover rate.

According to one embodiment of the present invention, the types of registration accounts may be such as demand deposit, savings account, loan, fund, insurance, stock, or bond.

According to an embodiment of the present invention, the credit risk analysis server 100 acquires total account information of an individual and a medium-sized company from a financial institution server for credit risk measurement based on an integrated account, and obtains an account number Alternatively, the accounts can be automatically classified according to the product type using the user input information inputted by the user.

The account information is illustratively shown in Fig. Referring to FIG. 3, the account number includes a point number-subject-serial number-verification number, and information described in the 'subject' can be used to classify the type of a product (account).

3, for example, in the 11-digit account number system, if the number listed in the course is 13 or 33, it means ordinary deposit, 18 or 38 means savings deposit, 26 means household check, 22 means corporate free deposits.

'Point number' means the branch code of the relevant financial institution issuing the relevant financial account.

The credit risk analysis server 100 can automatically classify the account information acquired from the financial institution server by the product type, using the account number and the user input information.

Referring to FIG. 9, the types of goods according to the embodiment of the present invention are illustrated as an example, and the credit risk analysis server 100 classifies and consolidates accounts by type and generates consolidated account information for each type. For example, the credit risk analysis server 100 may include integrated account information on demand deposit, integrated account information on a savings account, integrated account information on a loan, integrated account information on a fund, integrated account information on insurance, The integrated account information for the bonds, and the integrated account information for the bonds.

For example, the integrated account information for the demand deposit may include a consolidated account number as a virtual account number, a sum amount for deposits classified as a demand deposit, a withdrawal date and withdrawal amount, a deposit date, and a deposit amount have. As of September 21, 2015, a user has deposited 5 million won in ordinary savings account and 3 million won in ordinary deposit account at Bank B on September 21, 2015, and deposits 1,000,000 won into ordinary savings account of Bank A , And it is assumed that it is withdrawn from the ordinary bank account of Bank B, April 1, 2015 by 200,000 won. In this case, the consolidated account information for the demand deposit may include the following information.

Consolidated account number for demand deposit credit When to withdraw Amount withdrawn When to deposit Deposit amount **** - ** - ** - ** 8 million won 2015.4.1 200,000 won January 3, 2015 1 million won

These types of unified account information are illustrative and can be configured in other types. For example, as shown in the table of FIG. 10, the integrated account information for each type can be configured to include information on the original account number. 10 will be described later in detail.

Meanwhile, in the same or similar manner, the integrated account information for other types may also include information such as a consolidated account number, deposit amount, withdrawal amount, withdrawal amount, deposit time, and deposit amount.

The credit risk analysis server 200 can identify the cash flow for each type through the type of integrated account information and can determine which type of cash flow is worse or better.

Furthermore, the credit risk analysis server 200 may integrate the consolidated account information for each type to generate the entire consolidated account information. It is possible to integrate consolidated account information by type, to give a new consolidated account number, and to generate total consolidated account information that can determine the entire flow of cash.

As such, the credit risk analysis server 100 can integrate and manage accounts by type, and can manage withdrawals and deposits in real time by type.

Types of accounts can be like demand deposits, savings deposits, loans, funds, insurance, stocks, or bonds.

As will be described later, according to an embodiment of the present invention, the credit risk analysis server 100 classifies the entire financial accounts according to the product type and calculates the credit risk for each product type. The reason for calculating the credit risk for each product type is as shown in [Table 2] of FIG. 9, since it is possible to judge the transition phenomenon of the credit risk among the product types when the withdrawal and repurchase attributes by product type are utilized to be.

[Table 2] in Fig. 9 shows data on the withdrawal or repurchase characteristics of individual financial account receivable accounts among the statistics surveyed and announced by the parent bank in July 2013. According to these data, customers have a statistical feature that withdraw funds from demand deposit accounts first, rather than savings accounts, when funds are needed. The reason is that even if there is a financial difficulty, it means withdrawing the cash from the ordinary account book while keeping the savings account that you joined for the future. For example, demand deposits include checking accounts, ordinary deposits, quasi-deposits, and household comprehensive deposits. Savings deposits may include term deposits, periodic deposits, savings deposits, and free savings deposits.

The credit risk analysis server 100 automatically classifies the account type for each product type and the integrated account and generates the integrated account information on the basis of the virtual account code .

FIG. 5 is a diagram for explaining a method for generating consolidated account information by type according to an embodiment of the present invention.

Referring to FIG. 5, a method for generating integrated account information by type according to an embodiment of the present invention includes extracting a financial institution code / account number from account information; Extracting a pattern of an account number using a financial institution code and an account number; Checking whether the corresponding pattern exists in the account number mast DB; Confirming whether an account subject code exists if the pattern exists in the account number mast DB; Extracting the account subject code information if the account subject code exists, processing the account subject as 'other' if the corresponding pattern does not exist in the account number master DB or if the account subject code does not exist; Generating an integrated account number for each type using the extracted account item code information and the bank code; And integrated account information for each type including the integrated account number for each type may be generated and stored.

Since the position of the account subject code differs depending on the pattern of the account number for each financial institution and the pattern is different, in the method of generating the type-integrated account information according to the embodiment of the present invention, Perform the step of grasping the pattern of numbers.

The method of generating consolidated account information for each type according to an embodiment of the present invention includes checking whether a consolidated account number for a corresponding type is generated before generating a consolidated account number for each type, The integrated account number may not be generated.

For example, assuming that there is one free deposit account in the bank A and one free deposit account in the bank B, applying the method of generating the integrated account information for each type according to the embodiment of the present invention I will explain.

The method for generating integrated account information by type according to an embodiment of the present invention extracts a financial institution code / account number from the free deposit account information of the bank A. Extracts a pattern (for example, account digits) of the account number using the extracted account number and the financial institution code, and determines whether or not the corresponding pattern exists by referring to the account number mast DB. If there is a pattern of the free bank account of the bank A in the account number mast DB, the location of the account code in the free bank account information of the bank A can be known. If the account code is extracted by referring to the location where the account code exists, the extracted account code will be referred to as 'free deposit', so that the integrated account number for the requested deposit is generated.

The method for generating type-specific integrated account information according to an embodiment of the present invention subsequently extracts a financial institution code / account number from the free deposit account information of the bank B. Extracts a pattern (for example, account digits) of the account number using the extracted account number and the financial institution code, and determines whether or not the corresponding pattern exists by referring to the account number mast DB. If there is a pattern of the free bank account of the bank B in the account number mast DB, the location of the account subject code in the free bank account information of the bank B can be known. If you extract the account code by referring to the location where the account code exists, you can see that the extracted account code means 'free deposit'. However, since the integrated account information including the integrated account number for the demand deposit is already generated from the deposit account of the bank A, the integrated account information for the bank B is not newly created, The deposit amount, withdrawal amount, withdrawal date and time, deposit amount, deposit date, etc. of the bank B's free deposit account are integrated into the consolidated account information.

The integrated account information generated by the method of generating the integrated account information for each type described with reference to FIG. 5 is illustrated in FIG.

Referring to FIG. 10, unified account information on demanded deposits is illustrated for each customer 1 and customer 2, respectively.

Referring to FIG. 10, if two customers hold three and four demand deposit accounts in three banks, respectively, the integrated account number for each product type as shown on the right side of the table is generated.

In FIG. 10, the financial institution code 002 (industrial bank), 003 (industrial bank) and 004 (national bank) are actual code information, and the underlined part in the account number means an account code corresponding to the ordinary deposit Actual code information. In the institution category, 1001 is a virtual code for displaying a bank in the present invention, and 101 in a product type is a virtual code for displaying a demand deposit.

The transaction details (deposit timing, deposit amount, withdrawal time, withdrawal amount) of the consolidated account information and the total deposit amount are not shown in FIG. 10, but transaction details of the individual account and the total deposit amount can be naturally added to the consolidated account information of FIG.

Now, a description will be given of a method in which the credit risk analysis server 100 calculates the credit risk index based on the integrated account information.

Although there are various credit risk indicators, in embodiments of the present invention, real-time default probability, real-time deposit turnover rate, and real-time safety factor of the integrated account information base can be used as a credit risk indicator.

To this end, in the present invention, in order to apply the phenomenon called the delinquency and the withdrawal question, which is a core concept in the definition of a general default, to the consolidated account information, the following concept will be newly defined.

1) Delinquent: (T) The point of integration account <0

                  = (T-1) The number of the point-in-time consolidated accounts + (T)

                     <(T) The amount of the withdrawal of the time integration account

    2) Number of doubts: Pr ((T) Integrating time integration account <0)

If the nature of the flow is that the enemy is a cumulative amount of the balance over a certain period of time, the balance at a specific point in time has a character of a stock.

In addition, it is effective to use the whole integrated account to explain the credit risk such as the counterparty's default, although the integrated account can use both the integrated account of the type and the whole integrated account (the account that integrates the integrated accounts by type). This is because if the deficit is deemed to be insufficient due to insufficient demand, the bankruptcy can be prevented by raising the necessary amount through withdrawal and redemption of savings account or other financial account to be. However, when measuring the credit risk using the deposit turnover rate, it is more meaningful to use the integrated account information by product type than the total consolidated account as the leading indicator.

Based on this conceptual definition, a method of calculating the default probability and the safety factor from the integrated account information of the credit risk analysis server 100 will be described.

In this embodiment, the credit risk analysis server 100 utilizes a stress-strength model that is widely used to measure the reliability of a machine or a device in the field of engineering. Here, in the Stress-Strength model, stress is an external factor such as temperature and current which causes a failure of a material, a component and a device. Strength refers to an external load Means the ability of materials, components and devices to perform satisfactorily without failures.

Therefore, when considering from the viewpoint of bankruptcy, the withdrawal amount of the unified account at the time point (T) can be defined as the stress at that point in time, and the unity of the unified account at the time point (T) can be defined as the strength.

Since the withdrawal amount and the relative value vary with time, they have the characteristics of a random variable, and as a result statistical analysis using the probability distribution becomes possible. In this case, for convenience of explanation, the distribution characteristics of the withdrawal amount and the enemy number are the normal distribution (( Normal Distribution).

As mentioned above, the definition of delinquency or withdrawal questions and defaults will be redefined into three types as shown in [Table 4] of Figure 11 by using the withdrawal amount of the consolidated account and the appropriate information.

1) If the withdrawal amount of the consolidated account is constant but the withdrawal amount at the time point (T) is greater than the point-in-time point (T)

2) If the receipt amount and the withdrawal amount increase to the same magnitude (T), but the point of departure does not change, but the amount of withdrawal at time (T)

3) As the amount of withdrawal increases more than the increase in receivable amount, the amount of withdrawal at time (T) becomes larger than that at time (T).

The credit risk analysis server 100 according to the embodiment of the present invention can measure the credit risk using the turn-over ratio of deposits.

Hereinafter, the credit risk method based on the deposit turnover rate will be described in detail.

The deposit turnover rate of a financial account is the average number of withdrawals of the funds compared to the average of the withdrawals of the account during the unit period divided by the average during the same period (see Equation 1 in FIG. 12).

     In general, the deposit turnover rate is widely used as an indicator of market capitalization or liquidity indicators such as bond yield or bill default rate. The high deposit turnover rate can be divided into the following three types.

First, although the balance is constant, it is a positive aspect that suggests that the economy will improve if the withdrawal is frequent during the unit period for consumption or investment,

Second, although the amount of withdrawal is constant, the decrease in the amount of deposits has greatly reduced the level of interest, which is a negative aspect,

Third, although the total amount of withdrawals increased, the balance of payments decreased.

Therefore, it is based on the second and third definition of the above-mentioned deposit turnover rate to evaluate the credit risk of an individual or a small and medium-sized enterprise by utilizing the deposit turnover rate. As can be seen from the definition of the deposit turnover rate, In the second step. This is different from the method of the present invention in that the current personal credit rating system (CSS) reflects the customer's credit, that is, the debt information, but does not reflect the receipt, that is, the asset information.

In addition, in the present embodiment, the credit risk analysis server 100 is characterized in that the deposit turnover rate is calculated and utilized for each product type as shown in [Table 2] of FIG. 9, ], There is a transfer of credit risk by financial product type.

In other words, if the demand deposit is preferred over the savings deposit and the reimbursement order, the transition of the credit risk can be analyzed using the integrated account information of each product type rather than the individual account or the whole consolidated account level. Real-time proactive monitoring is possible.

That is, as shown in [Table 5] of FIG. 13, if the urgent fund is needed at time T1, the turnover of the savings deposit is maintained at a certain level even if the turnover rate of the demand deposit increases. The reason is that even if there is a financial difficulty, it means withdrawing the cash from the ordinary account book while keeping the savings account that you joined for the future. However, it can be seen that if this situation persists, and the balance of demand deposits is insufficient, it will be inevitable to terminate savings deposits and fund the necessary funds. In this case, since the savings amount is not large in most cases, the slope of the savings account turnover tends to be very short in a short period of time.

This credit risk transfer phenomenon can not be grasped by the credit risk measurement model using only the credit information like the existing evaluation model, because the integrated account information for each product type includes the account information . In other words, the consolidated account information on savings deposits is a combination of savings bank accounts information of domestic and foreign financial institutions such as Kookmin Bank, Shinhan Bank and Woori Bank held by individuals or SMEs. Therefore, It is possible to know in advance that the effects of the shortage of funds due to the withdrawal of funds or the increase in the turnover rate of deposits are sequentially distributed to the financial institutions.

In order to increase the discriminative power of the credit risk using the deposit turnover rate of the integrated account information for each product type, it is effective to utilize the beta for the withdrawal amount and the time series data of the adversary during the unit period as in the present embodiment. Beta is the regression coefficient or slope of the regression equation, which is a good indicator of trends in time series data.

This is because the deposit turnover rate itself is a concept of a stock, so if the turnover rates of the two counterparties are the same, credit risk itself can not be judged. Instead, since the beta of the withdrawal amount and the number of the deficit at the latest point in the analysis are of the flow type, it is possible to increase the ability to discriminate the risk by considering this information together with the deposit turnover rate.

According to the above description, the deposit turnover scaling criteria for the credit risk evaluation based on the integrated account information can be constructed as shown in Table 6 of FIG. 14, for example.

[Table 6] in FIG. 14 shows only a deposit turnover rate of demanded deposit products among various financial product types as a simple example, and a smaller number indicates a higher credit risk.

In order to utilize the present invention in the future, it is necessary to calculate the deposit turnover rate and the beta as well as the classification of all the financial products of all the financial institutions in addition to the demand deposit products as described above, and to determine the credit risk by scaling them have.

Next, the credit risk analysis server 100 can measure the credit risk using, for example, the Probability of Default and the Safety Ratio.

Hereinafter, a credit risk measurement method using the Probability of Default and the Safety Ratio will be described.

[Table 7] shows an example of the default occurrence case 3 shown in [Table 3] of FIG. 19 in order to explain the relationship between the default probability and the safety factor of the present invention and the credit risk. For example, It is the data of the withdrawal amount and the number of the unified accounts created by performing simulation of bank account deposit and withdrawal for 7 accounts (3 deposits, 2 deposits, 2 loans).

As shown in the right side of [Table 7] in Fig. 15, there is a safety zone between the enemy number and the payout amount distribution. However, as time passes, (T), the safety zone disappears and the bankruptcy occurs.

Here, the safety zone or the safety factor means the ratio to the load (withdrawal) and the strength (the competitor) of the consolidated account. For example, if the amount of withdrawals (x) and the number of competitors (y) follow normal random variables,

In this case, the safety factor can be defined as shown in Equation 2 in FIG.

That is,

, Where m is the safety factor, x is the withdrawal amount of the consolidated account, y is the cumulative amount of the unit time of the consolidated account,

를 만족하는 정규분포확률변수이고,

는 x에 대한 평균이고,

는 x 에 대한 표준편차이며,x is

Is a normal distribution random variable satisfying the following equation

Is the mean for x,

Is the standard deviation for x,

를 만족하는 정규분포확률변수이고,

는 y에 대한 평균이고,

는 y에 대한 표준편차이다.y is

Is a normal distribution random variable satisfying the following equation

Is the mean for y,

Is the standard deviation for y.

와

가 1이면 68.3%, 2이면 95.5%, 3이면 99.7%에 해당한다.In the case of normal distribution,

Wow

Is 1, 68.3%, 2 is 95.5%, and 3 is 99.7%.

In addition to the normal distribution, other types of probability distributions such as Poisson distribution, lognormal distribution, weibull distribution, and beta distribution are possible.

The relationship between the safety factor and the default probability will be described in the following examples.

For example,

, The safety factor at the 95% confidence interval is as follows.

)이므로, 여기서 z = x-y라 두면 z는 마찬가지로 정규확률변수이므로, 그 분포특성은 도 12에서의 수학식 3과 같다. And, as we have seen above, if the amount of withdrawals at a particular point is greater than the appropriate number,

), So that if z = xy, z is also a normal random variable, and its distribution characteristic is expressed by Equation (3) in FIG.

In other words,

,

,

,

,

.

)은 도 12에서의 수학식 4와 같다. In the end,

) Is shown in Equation (4) in Fig.

In other words,

.

When the integration lower bound for the companies A and B in the above example is calculated,

A company's integration limit is

The integration limit of company B is

, And accordingly

          The probability of bankruptcy of company A is (1-NORMSDIST (3.535)) * 100 = 0.02%

          The probability of bankruptcy of company B is (1-NORMSDIST (0.624)) * 100 = 26.6%.

Here, NORMSDIST is a function to obtain the standard normal cumulative distribution value provided by MicroSoft's EXCEL program.

The scaling criteria of the default probability and the safety factor for the credit risk evaluation based on the integrated account information can be constructed as shown in Table 8 of FIG.

Referring to Table 8 in FIG. 16, the lower the order of the risk level, the higher the risk.

In the present invention, the deposit turnover rate, the default probability, and the safety factor are calculated as credit risk indicators using the integrated account information held by the individual or small and medium enterprises as described in the above embodiments. Further, Score), which can be determined by the current real-time credit risk of a particular individual or small business.

At this time, there are various criteria for calculating the total credit score, but a detailed description will be omitted in the embodiment of the present invention.

4 is a flowchart illustrating a credit risk management method based on integrated account information according to an embodiment of the present invention.

Referring to FIG. 4, a method for managing credit risk based on integrated account information according to an embodiment of the present invention includes classifying registered accounts by product type, generating consolidated account information for each product type, Calculating a probability, calculating a safety factor for the aggregated account, calculating a deposit turnover for the aggregated account, and calculating the aggregate credit score of the aggregated account.

The step of classifying the registered accounts according to the product type includes, for example, a step of registering / authenticating the credit risk management service based on the integrated account information, and a step of registering the financial institution / account number / authorized certificate, It is performed after registration.

The step of generating the integrated account information may be performed by extracting transaction information for each account through screen scraping based on the goods classified by the product type with respect to the registered account. Screen scraping refers to an operation of accessing a financial server to obtain transaction information (deposit amount, deposit / withdrawal transaction history and date / time) for each account.

The method of classifying classified accounts by product type and using the account numbers and the account number and generating the integrated account information has been described in detail with reference to FIGS. 5 and 10, and will not be described here.

The step of calculating the default probability of a consolidated account includes calculating the unit period number of the consolidated account, calculating the total withdrawal amount of the consolidated account, calculating the average and standard deviation of the total amount of the withdrawal and withdrawal, and the like . &Lt; / RTI &gt;

The step of calculating the safety factor for the consolidated account can be calculated based on the average and standard deviation of the total and withdrawal amounts.

The step of calculating the deposit turnover rate for the unified account can be calculated based on the unit period average balance calculation result of the unified account. On the other hand, the unit time average balance calculation result of the consolidated account can be calculated from the unit time withdrawal calculation result of the consolidated account.

Here, the method of calculating the bankruptcy probability, the safety factor, and the deposit turnover rate has been described with reference to FIGS. 12, 13, 15, and 16, and therefore will not be described here.

Each of the steps described with reference to FIG. 4 can be performed, for example, by the above-described credit risk analysis server.

6 is a flow chart illustrating an example of an exemplary process for generating a default probability and a security factor using aggregated account information.

6, an exemplary process for generating the default probability and the safety factor includes calculating an amount of withdrawal and an amount in a unit period of an account, calculating an average and standard deviation of the withdrawal amount, calculating an average and standard deviation of the enemy number, Calculating an average and standard deviation of the amount of withdrawal from the withdrawal amount, calculating a default probability, determining a statistical significance level, and calculating a safety factor.

7 is a flow chart illustrating an example of an exemplary process for generating deposit turnover using aggregated account information.

Referring to FIG. 7, an exemplary process for generating deposit turnover comprises: extracting an amount of withdrawal and an amount in a unit period of an account, calculating a withdrawal amount and a withdrawal amount beta, calculating a total amount of the total amount, , And calculating a deposit turnover rate of the account.

FIG. 8 shows an exemplary process for generating a total credit score of a counterparty using a default probability, a safety factor, and a deposit turnover rate.

8 shows an exemplary process for generating a total credit score (Score), which may include the steps shown in FIG. That is, the steps for extracting the safety factor score (extracting safety factor of integrated account, mapping to safety factor scaling DB, extracting safety factor score), steps for extracting turnover score (Step of extracting turnover rate, step of mapping to turnover rate scaling DB, step of extracting turnover rate score), steps for extracting a bankruptcy probability score (step of extracting bankruptcy probability of unified account, step of mapping bankruptcy probability scaling DB, and calculating a total credit score from the safety factor score, the turnover score, and the default probability score.

On the other hand, with respect to the contents described or shown in the drawings of the present application, parts not mentioned in the detailed description of the present application are incorporated as a part of this specification.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the invention. It is therefore to be understood that the embodiments of the invention described herein are to be considered in all respects as illustrative and not restrictive, and the scope of the invention is indicated by the appended claims rather than by the foregoing description, Should be interpreted as being included in.

10: Network
100: Credit Risk Analysis Server
200: Member terminal
300: Server of financial institution

Claims (12)

On the computer,
Classifying the registered accounts by type;
Generating consolidated account information for registered accounts classified by type;
Calculating a plurality of credit risk indices for the unified account based on the unified account information; And
And calculating a total credit score from the credit risk indices based on the credit risk indexes. The computer-readable recording medium records a program for executing a real-time credit risk analysis method based on a unified account information. The method according to claim 1,
Wherein the credit risk indices include at least one of a safety factor, a default probability, and a deposit turnover rate. The method according to claim 1,
The types of registration accounts include,
A demand deposit, a savings deposit, a loan, a fund, an insurance, a stock, a stock, and a bond. 3. The method of claim 2,
The step of calculating the credit risk indices comprises:
Calculating a deposit turnover rate for each type of registered account,
Wherein the turnover rate is defined as a value obtained by dividing the total withdrawal amount of an account for a unit time by a flat rate for the same period. 3. The method of claim 2,
The safety factor,

, Where m is the safety factor, x is the withdrawal amount of the consolidated account, y is the cumulative amount of the unit time of the consolidated account,
x is

Is a normal distribution random variable satisfying the following equation

Is the mean for x,

Is the standard deviation for x,
y is

Is a normal distribution random variable satisfying the following equation

Is the mean for y,

Is a standard deviation of y. &Lt; RTI ID = 0.0 &gt; 11. &lt; / RTI &gt; 6. The method of claim 5,
The default probability

Lt; / RTI &gt;

,

,

,

The program being recorded on a computer-readable recording medium. In a real-time credit risk analysis system based on integrated account information,
A server of a financial institution that stores information about registered accounts; And
A plurality of credit risk indices are calculated for the unified account based on the unified account information, and the credit risk indices are calculated based on the unified account information, And a credit risk analysis server for calculating a total credit score based on the integrated credit information. 8. The method of claim 7,
Wherein the credit risk indices include at least one of a safety factor, a default probability, and a deposit turnover ratio. 8. The method of claim 7,
The types of registration accounts include,
Real-time credit risk analysis system based on integrated account information, which is composed of demand deposit, savings deposit, loan, fund, insurance, stock, stock, and bonds. 8. The method of claim 7,
The operation of calculating the credit risk indices
Calculating a deposit turnover rate for each type of registered account,
Wherein the turnover rate is defined as a value obtained by dividing the total withdrawal amount of an account for a unit time by a flat rate for the same period. 9. The method of claim 8,
The safety factor,

, Where m is the safety factor, x is the withdrawal amount of the consolidated account, y is the cumulative amount of the unit time of the consolidated account,
x is

Is a normal distribution random variable satisfying the following equation

Is the mean for x,

Is the standard deviation for x,
y is

Is a normal distribution random variable satisfying the following equation

Is the mean for y,

Is a standard deviation for y. &Lt; RTI ID = 0.0 &gt; A &lt; / RTI &gt; 9. The method of claim 8,
The default probability

Lt; / RTI &gt;

,

,

,

Wherein the credit risk management system comprises:

Images