JP2017182284A - Loan examination support device and system, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To support more meticulous loan examination by using data indicating a relationship between a financial institution and a client, that is data which has been stored internally within the financial institution and not effectively used up to now, and thereby increase a ratio of proper lending, reduce guarantee charges and increase revenue of the financial institution.SOLUTION: A loan examination support device includes: a score determination section which calculates a default rate of a client in order to determine a rank; and a credit determination section for determining whether proper lending is possible or not. The score determination section selects at least one model from among plural models expressing a relation between loan application information with financial institution internal information and a score for calculating a default rate of the client, and then calculates a default rate of the client from the selected model on the basis of the loan application information and/or the financial institution internal information related to the client in order to determine a rank. The credit determination section determines whether proper lending is possible or not with regard to the applied loan on the basis of at least the rank.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a system or the like that supports examination processing for loan application input including various loans in a financial institution, and more particularly to a system or the like that supports credit judgment processing at the time of examination.

  A loan credit determination system that automatically performs loan acceptance, credit, loan execution, and the like is known (Patent Document 1).

  Patent Document 1 provides a loan credit judgment system that uses logistic regression analysis to create a loan acceptance / rejection judgment standard and automatically corrects the acceptance / rejection judgment standard based on the latest delinquent credit loss state.

  Specifically, the loan credit determination system described in Patent Document 1 includes an attribute classification creating unit that performs customer attribute classification based on input information, and a related factor that sets a related factor according to the input information for each attribute classification Factor setting means; regression coefficient setting means for setting a regression coefficient corresponding to the related factor for each attribute classification; and selecting the corresponding attribute classification based on customer information and determining an appropriate rate for the relevant factor. An arithmetic unit that performs arithmetic processing, and an approval unit that approves a loan if the arithmetic result of the arithmetic unit is a preset appropriate rate, the factors include customer attribute information, real estate information, funds Information, family information, personal credit information, repayment plan information, and history information. Points are calculated based on these information, and the credit of the borrower is automatically Divide.

JP 2008-90689 A

  As shown in FIG. 16, the loan contracted interest rate, the guarantee fee paid to the guarantee company, and the default rate are generally linked. The amount obtained by subtracting the guarantee fee and default rate from the contracted interest rate of the loan is roughly the profit that the financial institution can obtain. Therefore, it is expected that the profit of financial institutions will be improved by reducing the guarantee fee or default rate.

  Therefore, according to the present invention, in loan examination, a loan that has been treated as a guaranteed loan (a loan for requesting a guarantee from a guarantee company) in the past is a proper loan (a financial institution takes a risk without requesting a guarantee from the guarantee company). An object of the present invention is to provide a loan examination support device that supports processing as a loan.

The present invention includes the following aspects.
[1]
A score determination unit that calculates the default rate of the customer and determines the rank;
A loan examination support device comprising a credit judgment unit for judging whether proper loan for an application loan is possible,
The score determination unit
A plurality of models representing relationships between loan application information and / or internal financial institution information about customers and scores used to calculate customer default rates, each of which has a different relationship among the models To select at least one model based on internal financial institution information about the customer,
Based on the loan application information and / or financial institution internal information about the customer, the default rate of the customer is calculated from the selected model,
A rank is determined based on the calculated default rate,
The credit judgment unit
Determining whether proper financing for the application loan is possible based on at least the rank;
The loan examination support apparatus characterized by the above.
[2]
For example, the model selected by the score determination unit when the deposit information is included in the internal information of the financial institution related to the customer is selected when the score is assigned to the deposit information of the customer and there is no deposit information. The loan examination support apparatus according to [1], wherein the model does not have the rating assigned to customer deposit information.
[3]
For example, when there is past loan information in the financial institution internal information about the customer, the score selected by the score determination unit is assigned the score for the customer's past loan information, and there is no past loan information The loan examination support apparatus according to the above [1] or [2], wherein the selected model is not assigned to the customer's past loan information.
[4]
Examples of loan application information and / or internal information about financial institutions related to customers include, for example, whether there are deposit accounts, deposit account opening dates, deposit history, account transaction history, deposit balance, salary or pension transfer, and public charges Withdrawal, customer's annual income, family structure, workplace industry, employment style, job title, length of service, residence classification, housing type, loan age, past loan history, customer to the financial institution Types of loans that have been applied in the past, number of applications, number of loans, amount of loans, repayment amount, repayment status, accident information related to repayment of loans, and personal credit information, etc. The loan examination support device according to any one of the above [1] to [3], which includes at least one of (not limited list).
[5]
A foreign credit judgment unit that judges whether personal credit information about customers provided by an external credit agency contains information on a serious financial accident;
A loan amount determination unit that calculates the annual income ratio and repayment ratio of the customer based on the loan application information and / or the internal information of the financial institution relating to the customer;
The credit determination unit determines whether proper loan for the application loan is possible based on the rank, the determination result by the foreign credit determination unit, and the annual income ratio and the repayment ratio calculated by the loan amount determination unit. The loan examination support device according to any one of the above [1] to [4].
[6]
A score determination unit that calculates the default rate of the customer and determines the rank;
A loan examination support device comprising a credit judgment unit for judging whether or not a guaranteed loan being executed by a financial institution can be changed to a proper loan,
The score determination unit
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Select at least one model based on information,
Based on the financial institution internal information about the customer, calculate the default rate of the customer from the selected model,
A rank is determined based on the calculated default rate,
The credit judgment unit
Determining whether the guaranteed loan can be changed to a proper loan based on at least the rank;
The loan examination support apparatus characterized by the above.
[7]
A score determination unit that calculates the default rate of the customer and determines the rank;
A loan examination support device comprising a credit judgment unit that judges whether or not the credit limit of a customer should be reduced in a credit in progress of a loan being executed by a financial institution,
The score determination unit
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Select at least one model based on information,
Based on the financial institution internal information about the customer, calculate the default rate of the customer from the selected model,
A rank is determined based on the calculated default rate,
The credit judgment unit
The rank of the customer determined by the score determination unit at the time of entrusting credit has improved to a rank higher than a predetermined rank number compared to the rank of the customer determined by the score determination unit at the time of loan examination If it is determined that the credit line of the customer should be expanded, or if it has deteriorated to a rank lower than a predetermined number of ranks, it is determined that the credit line of the customer should be reduced,
A loan examination support device characterized by that.
[8]
A loan examination support system including the loan examination support apparatus according to any one of [1] to [7].
[9]
A score determination step of calculating a default rate of the customer and determining a rank;
A program for causing a computer to execute a credit judgment step for judging whether or not a proper loan for an application loan is possible,
The score determination step includes
A plurality of models representing relationships between loan application information and / or internal financial institution information about customers and scores used to calculate customer default rates, each of which has a different relationship among the models Selecting at least one model based on internal financial institution information about the customer,
Calculating a customer default rate from the selected model based on loan application information and / or financial institution internal information about the customer; and determining a rank based on the calculated default rate;
Including
The credit determination step includes
A program comprising the step of determining whether or not proper financing for the application loan is possible based on at least the rank.
[10]
For example, the model selected by the score determination process when there is deposit information in the internal information of the financial institution relating to the customer is selected when the score is assigned to the customer's deposit information and there is no deposit information. The program according to [9], wherein the model does not have the rating assigned to customer deposit information.
[11]
For example, when there is past loan information in the financial institution internal information related to the customer, the model selected by the score determination process is assigned the score for the customer's past loan information, and there is no past loan information. The program according to [9] or [10], wherein the selected model is not assigned to the customer's past loan information.
[12]
Examples of loan application information and / or internal information about financial institutions related to customers include, for example, whether there are deposit accounts, deposit account opening dates, deposit history, account transaction history, deposit balance, salary or pension transfer, and public charges Withdrawal, customer's annual income, family structure, workplace industry, employment style, job title, length of service, residence classification, housing type, loan age, past loan history, customer to the financial institution Types of loans that have been applied in the past, number of applications, number of loans, amount of loans, repayment amount, repayment status, accident information related to repayment of loans, and personal credit information, etc. The program according to any one of [9] to [11], including at least one of (not limited list).
[13]
Foreign credit judgment processing to determine whether personal credit information about customers provided by external credit agencies contains information on serious financial accidents,
A loan amount determination process for calculating a customer's annual income ratio and repayment ratio based on loan application information and / or internal financial institution information about the customer,
The credit determination process determines whether proper loan for the application loan is possible based on the rank, the determination result by the foreign credit determination process, and the annual income ratio and the repayment ratio calculated by the loan amount determination process. The program according to any one of [9] to [12] above.
[14]
A score determination process that calculates the default rate of the customer and determines the rank;
A program for causing a computer to execute a credit determination process for determining whether a guaranteed loan being executed by a financial institution can be changed to a proper loan,
The score determination process includes
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Selecting at least one model based on information,
A process for calculating a default rate of a customer from the selected model, based on internal information about the financial institution relating to the customer, and a process for determining a rank based on the calculated default rate;
Including
The credit determination process includes
A process of determining whether the guaranteed loan can be changed to a proper loan based on at least the rank;
The program characterized by including.
[15]
A score determination process that calculates the default rate of the customer and determines the rank;
A program for causing a computer to execute a credit determination process for determining whether or not to reduce a credit line of a customer in a credit in progress of a loan being executed by a financial institution,
The score determination process includes
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Selecting at least one model based on information,
A process of calculating a default rate of the customer from the selected model based on internal information about the financial institution relating to the customer, and a process of determining a rank based on the calculated default rate,
The credit determination process includes
The rank of the customer determined by the score determination unit at the time of entrusting credit has improved to a rank higher than a predetermined rank number compared to the rank of the customer determined by the score determination unit at the time of loan examination The credit line of the customer is determined to be expanded, or the process of determining that the credit line of the customer should be reduced if the customer's credit limit is deteriorated to a rank lower than a predetermined number of ranks. Program.
[16]
The processes for calculating the default rate and determining the rank are performed for each of a plurality of types of loans, and the credit determination process is performed for each of the plurality of types of loans based on the loan application information. The device according to [1] above.
[17]
The default rate calculation process and the rank determination process are performed for each of a plurality of types of loans, and the credit determination process is performed for each of the plurality of types of loans based on the loan application information. [9] The program according to [9] above.
[As a further improved device or system]
A system including a marketing support device or a device including a score determination unit that calculates a default rate of a customer and determines a rank,
The score determination unit
A plurality of models representing a relationship between internal information of a financial institution relating to a customer and a score used to calculate a default rate of the customer, and at least one model is selected from the plurality of models each having the different relationship And
Based on the financial institution internal information about the customer, calculate the default rate of the customer from the selected model,
Determine the rank of the customer based on the calculated default rate,
A marketing support apparatus or a system including the apparatus, wherein customer information having a rank higher than a predetermined rank is output to a terminal in association with the rank of the customer.

In the loan examination support apparatus according to an embodiment of the present invention, data that has been accumulated in the financial institution but has not been used effectively so far (hereinafter referred to as “internal data of the financial institution” or “internal information of the financial institution”). By using data that shows the relationship between financial institutions and customers, and changing the screening (score) model for calculating the default rate of customers, more detailed loan screening is possible. . In addition, the relationship between the rank determined by the loan examination support device and the expected margin (expected profit) becomes clear, and it takes an excessive risk to apply for a loan that was previously determined to be a guaranteed loan. It is possible to undertake as a proper loan. As a result, the guarantee fee paid to the guarantee company is reduced, and the profit of the financial institution is improved.
More specifically, or in other words, the following effects are also achieved.
(1) Since the guarantee company cannot access the data stored in the financial institution in the first place, the loan examination according to the embodiment of the present invention can be detected conventionally by effectively using the data. It is possible to pursue further expected profits that did not exist.
(2) Although a plurality of the above-described models are prepared (described later), these models occupy a very important position in the credit determination processing according to an embodiment of the present invention. These review (score) models can be designed and scored by freely combining necessary variables and conditions in data including internal data of financial institutions according to the required specifications of the model. It can be incorporated into a system according to an embodiment of the present invention as a model.
(3) The internal data of the financial institution utilized for constructing a plurality of examination (score) models in the system according to an embodiment of the present invention is also utilized for model reconstruction (review and improvement). It can also be used for revenue forecast calculation and marketing (described later).

It is a schematic diagram of a loan examination support system in one embodiment. It is a functional block diagram of the loan examination support system in one embodiment. It is a flowchart of loan examination in one embodiment. It is a flowchart of the external shape reference | standard determination process in one Embodiment. It is a flowchart of the temporary score determination process in one Embodiment. It is a flowchart of the model selection process in one Embodiment. It is a flowchart of the default rate calculation process in one embodiment. It is an example of the scoring table of the model 1 in one Embodiment. It is an example of the scoring table of the model 2 in one Embodiment. It is an example of the scoring table of the model 3 in one Embodiment. It is an example of the criteria of rank in one embodiment. It is a flowchart of the proper credit determination process in one embodiment. It is an example of the proper credit criterion in one Embodiment. It is a table | surface which shows an example of the expectation sheath (expected profit) in one Embodiment. It is a table | surface which shows an example of the expectation sheath of the credit on the way in one Embodiment. It is a graph which shows the relationship of the conventional general contracted interest rate, a guarantee fee rate, and a default rate.

  Embodiments for carrying out the present invention will be described below in detail with reference to the drawings.

[First Embodiment]
FIG. 1 is a schematic configuration diagram of a loan examination support system 100 including a loan examination support apparatus 10 according to the first embodiment of the present invention.

  The loan screening support system 100 includes a loan screening support device 10, a foreign trust determination device 13, a database 20 that stores internal information of financial institutions, and a screening terminal 30 that is used for final loan approval / disapproval (acceptance or rejection). included.

  Here, in this specification, financial institutions include banks, credit unions, cooperatives, insurance companies, consumer finance, or government-affiliated financial institutions. Loan guarantee companies are not included in financial institutions. Loans include various types of loans such as free loans for individuals, purpose loans (education funds, car purchases, renovations, etc.), card loans, mortgages, cashing, and corporate loans. External credit institutions (hereinafter also referred to simply as “credit institutions”) include CIC Co., Ltd. (CIC), Japan Bank Personal Credit Information Center (KSC), and Japan Credit Information Organization (JICC). At least one institution is included.

  In this specification, the terms “loan”, “loan”, “borrow”, “loan”, and the like are used synonymously with the term “loan”. The term “application loan” refers to a loan that a customer has applied for. For example, “proper loan” for “application loan” means that the loan applied by the customer is proper (without a guarantee company). Indicates that a loan is possible. In the following description, a bank is assumed as a financial institution for easy understanding of the invention, but the present invention is not limited to this.

  An example of the hardware configuration of the loan examination support apparatus 10 is a computer (server) including a CPU, a main storage unit, an auxiliary storage unit, and an input / output unit (not shown). As shown in FIG. 1, the loan examination support apparatus 10 is communicably connected to the foreign communication determination apparatus 13, the database 20, and the examination terminal 30. The loan examination support apparatus 10 may exchange data with the terminal 50 of an external guarantee company using a file transfer tool (for example, HULFT (registered trademark)). Alternatively, when the loan examination support apparatus 10 determines that the guarantee is a guarantee request from the guarantee company in S508 or S1211 described later, the person in charge of the financial institution sends a document relating to the guarantee request to the guarantee company by FAX (or mail) for guarantee. You may ask the company to guarantee the loan.

  A hardware configuration example of the foreign communication determination device 13 is a computer (server) including a CPU, a main storage unit, an auxiliary storage unit, and an input / output unit (not shown). The foreign trust judging device 13 is communicably connected to the terminals 40 of a plurality of external credit institutions, collects and analyzes personal credit information provided from the plurality of external credit institutions, and analyzes the analysis results (foreign trust judgment processing results). This is provided to the loan examination support device 10. One embodiment of the foreign communication judging device 13 is a server equipped with a personal credit information inquiry system “e-Acris III plus” (manufactured by Profit Cube Inc.). It should be noted that the loan examination support device 10 has a function of foreign communication determination processing (foreign communication determination unit) by the foreign communication determination device 13 and the loan examination support device 10 (external communication determination of the loan examination support device 10 without using the foreign communication determination device 13). May communicate directly with the terminal 40 of the external credit agency.

  The examination terminal 30 in the financial institution may be communicably connected to a branch office terminal 35 disposed at another sales office of the financial institution. Moreover, you may comprise so that it can apply for a loan with a financial institution from terminal 60, such as a customer's PC or a smart phone.

  FIG. 2 is a functional block diagram of the loan examination support device 10, foreign communication judgment device 13, database 20, examination terminal 30, and external credit agency terminal 40.

  As shown in FIG. 2, the loan examination support device 10 includes, as function units realized by a CPU based on a program or software stored in an auxiliary storage unit or the like, an outer shape reference determination unit 11, a provisional score determination unit 12, A score determination unit 14, a loan amount determination unit 15, and a credit determination unit 16 are included. As described above, the loan examination support device 10 may include a foreign communication determination unit.

  In addition, the loan examination support apparatus 10 stores, in its auxiliary storage unit or the like, customer loan application information provided from the examination terminal 30 (referred to as “application information”) and various types of financial institution stored in the database 20. A scoring table 17 for converting information into a score (score) for calculating a default rate (referred to as “DF rate”) is stored. The scoring table 17 may be stored in the database 20, and the loan examination support apparatus 10 may access the database 20 and refer to the scoring table 17.

  The loan examination support device 10 can access the database 20 and use the internal information of the financial institution stored in the database 20 for credit judgment processing at the time of loan examination. The database 20 stores deposit information, customer personal information, past loan information, and other information that contributes to loan examination as internal information of the financial institution. As a premise of this specification, an external guarantee company cannot access internal information of financial institutions.

  In one embodiment, the “deposit information” includes presence / absence of a deposit account, opening date of a deposit account, deposit history, account transaction history, deposit balance, presence / absence of salary or annuity transfer, presence / absence of utility charges, and financial accident This includes the presence or absence of. “Customer information” includes the customer's name, age, address, annual income, work place information, family structure, presence / absence of collateral, and the like. The customer's work location information includes information such as employment type such as regular / non-regular or self-employed, job type, job title at the workplace, the number and type of employees of the workplace, and the number of months of service. The customer family configuration information includes information such as the presence / absence of the customer's family, the number of family members, and the presence / absence of children. The collateral holding presence / absence information includes information such as the type of residence of the customer such as a home / rental house, a detached house / apartment, and a residence classification. “Past loan information” includes the types of loans and the number of applications that the customer has previously applied to the financial institution, past loan history, loan count, loan amount, unpaid amount, repayment status, repayment delay, etc. Information, and personal credit information of customers acquired and stored in the past from external credit agencies.

  Further, the loan examination support device 10 obtains customer loan application information from the examination terminal 30 and obtains a determination processing result (foreign communication determination processing result) related to the customer's personal credit information from the foreign communication determination device 13. The application information includes the type of loan desired by the customer, the borrowing amount, the repayment period, the customer's name, age, address, telephone number, annual income, company information, and borrowing status at the relevant financial institution or other financial institution. Information that should be declared when receiving a loan. The personal credit information provided from an external credit institution includes information such as customer credit information, payment information, and serious financial accident information such as transfer and repayment delay.

  FIG. 3 is a flowchart showing an outline of loan examination by a financial institution.

  First, in S301, the financial institution inputs loan application information from the customer terminal 50 or the branch office terminal 35. At this time, the loan examination support device 10 (or examination terminal 30) assigns the applicant identification number to the customer at this stage, except for the customers who have already been given the applicant identification number. This makes it possible to identify a customer who refuses a loan application at the time of a subsequent loan application. The loan application may be a written application by FAX or mail, but in that case, it is desirable that the contents of the application information be converted into electronic data by a financial institution.

  Thereafter, the loan screening support apparatus 10 sequentially performs an outline reference determination process in S302, a provisional score determination process in S303, and a proper credit determination process in S304. Details of each process of the outline reference determination process, the provisional score determination process, and the proper credit determination process will be described later. Here, in the proper credit determination process, it is determined whether proper financing is possible. Note that the outer shape reference determination processing in S303 may be omitted according to the required specifications.

  In S305, based on the result of the proper credit determination process output by the loan screening support apparatus 10, the screening staff determines whether or not the loan execution is approved. After the approval of the result of the approval / disapproval determination by the examination supervisor is obtained in S306, the customer is notified of the result of the approval / disapproval of the loan in S307. In this way, whether or not to accept the loan application that the customer applied for is determined. In addition, you may abbreviate | omit the process of S305-S306 according to the intention of a financial institution.

  Next, with reference to FIGS. 4-12, each process of the external form reference | standard determination process (S302) by the loan screening assistance apparatus 10, a temporary score determination process (S303), and a credit determination process (S304) is demonstrated.

<Outline reference judgment processing> (Can be omitted depending on required specifications)
First, in the outline standard determination process (S302), the loan examination support device 10 calculates the annual income ratio and the repayment ratio from the application information provided by the customer, and the calculated annual income ratio and the repayment ratio exceed the predetermined standards, respectively. Judge whether or not. FIG. 4 is a detailed process flowchart of the outer shape reference determination process.

  In FIG. 4, in step S <b> 401, the outline reference determination unit 11 of the loan examination support device 10 acquires application information from the examination terminal 30, and as one embodiment, the customer's annual income, the desired borrowing amount of the loan to be applied, and other existing ones. Extract information such as loan amount and repayment years.

  In step S <b> 402, the outer shape reference determination unit 11 calculates a “repayment ratio” (%) based on the formula “total annual repayment amount ÷ annual income”. If the customer has other borrowings (including other financial institutions), the total annual repayment amount is also added. In step S <b> 403, the outline reference determination unit 11 determines whether the calculated “repayment ratio” is equal to or less than a predetermined reference value. The reference value is appropriately set according to the required specification, and may be, for example, 25% to 60%.

  In step S <b> 404, the outline standard determination unit 11 calculates “annual income ratio” (%) based on the expression “total borrowing amount / annual income amount”. The total borrowing amount includes the borrowing amount if the customer has other borrowings (including other financial institutions). In step S <b> 405, the outline reference determination unit 11 determines whether the calculated “annual income ratio” is equal to or less than a predetermined reference value. The reference value is appropriately set according to the required specification, and may be, for example, 35% to 80%.

  When both the “repayment ratio” and the “annual income ratio” are equal to or less than the reference value (Yes in S403 and S405), the process proceeds to the next provisional score determination process (S303).

  If the “repayment ratio” exceeds the reference value (No in S403) or if the “annual income ratio” exceeds the reference value (No in S405), the outer shape reference determination unit 11 rejects the loan application to the examination terminal 30 in S406. The loan examination by the loan examination support device 10 is finished. Upon receiving the notification, the examination terminal 30 displays on the display that the loan application should be rejected, and informs the person in charge of examination. Even when a notice to the effect that the loan application should be rejected is received from the loan examination support device 10 at the judgment of the person in charge of examination and the person in charge of examination (No in S403 or No in S405), the next provisional score You may make it progress to determination processing (S303).

<Tentative score determination process>
Next, in the provisional score determination process (S303), it is determined whether or not the loan application is such that a proper loan (a loan with no guarantee company attached and a financial institution taking a risk) is possible after the outline reference determination process. Thereby, the financial institution can reject the loan application before making an inquiry to the external credit institution (foreign communication inquiry) for an application that is obviously impossible to lend. FIG. 5 is a detailed process flowchart of the provisional score determination process.

  In FIG. 5, the provisional score determination unit 12 of the loan examination support apparatus 10 acquires application information from the examination terminal 30 and extracts information used for provisional score determination processing in S501. In S502, the provisional score determination unit 12 accesses the database 20, acquires internal information about the financial institution regarding the customer who has applied for the loan, and extracts information used for the provisional score determination process.

  In S503, the provisional score determination unit 12 selects a model used for calculating the DF rate based on the internal information of the financial institution (or both the internal information of the financial institution and the application information). In the present invention, the difference between models is a difference in calculation formulas used to calculate the DF rate (difference in scoring tables described later). In S504, the provisional score determination unit 12 calculates the customer DF rate in the application loan based on the model selected in S503. Here, the customer's DF rate in the application loan is the customer's DF rate in the case of a loan applied by the customer (desired borrowing amount, repayment years, loan type, etc.). In S505, the provisional score determination unit 12 determines a rank based on the DF rate calculated in S504. Details of the processes S503 to S505 will be described later.

  In S506, the provisional score determination unit 12 determines whether or not the rank determined in S505 is within a predetermined range. The predetermined range may be a range equal to or higher than a certain rank threshold (for example, a range of R05 or higher shown in FIG. 11 described later) or a specific range (for example, a range of ranks R03 to R06 shown in FIG. 11 described later). ). If the rank is within the predetermined range (Yes in S506), the provisional score determination unit 12 determines in S507 that the process may proceed to a proper credit determination process, and the process proceeds to S304 (FIG. 3). If the rank is not within the predetermined range (No in S506), in S508, the provisional score determination unit 12 transmits information (inquiry data) related to the customer's loan application to the terminal 50 of the guarantee company, and the loan to the guarantee company. Make a warranty request.

  The predetermined standard in S506 may be changed according to the type of loan and the expected margin (expected profit) obtained by the financial institution. For example, FIG. 14 is a table showing an example of the relationship among ranks R01 to R10, estimated annual DF ratio, guarantee fee, credit cost, and annual expected sheath obtained by analyzing financial institution internal information of financial institutions. is there. FIG. 14A shows the case of a target loan, and FIG. 14B shows the case of a free loan. In the table shown in FIG. 14, “guarantee fee” is a cost (cost) generated by requesting a guarantee from the guarantee company, and “credit cost” is a financial institution when the guarantee of the guarantee company is stopped. Is the potential cost (cost) incurred by taking the risk. In addition, “expected sheath” in the same table indicates an expected profit value generated when the guarantee fee is set to zero. For example, in the case of the target loan shown in FIG. 14A, the expected pod is positive for ranks R01 to R05, and the expected pod is negative for ranks R06 to R09. Therefore, in the case of the target loan, the reference rank in S506 may be set as R05. In the case of the free loan shown in FIG. 14B, the expected sheath is positive for ranks R01 to R09, and negative for rank R10. Therefore, in the case of the free loan, the reference rank in S506 may be set as R08 or R09.

  Details of the model selection process (S503) used for calculating the DF rate will be described with reference to FIG. In this embodiment, there are three types of models for calculating the DF rate, but the present invention is not limited to this. In the present invention, the number of models for calculating the DF rate may be two or more, and may be two or more. The model selection flow shown in FIG. 6 is an example, and the present invention is not limited to this example.

  In FIG. 6, first, in S601, the provisional score determination unit 12 determines, for example, whether there is customer deposit information (existence of a deposit account, date of deposit account opening, account transaction history, etc.) from financial institution internal information. Although not limited, when the customer's account is opened before the month before the loan execution date, the provisional score determination unit 12 determines that the deposit information is “present” (Yes in S601), and if not, May be determined as “none” (No in S601).

  In step S602, the provisional score determination unit 12 determines whether there is customer's past loan information from the internal information of the financial institution, for example. Although not limited, when the transaction history of another loan exists in the financial institution before the loan execution date, the provisional score determination unit 12 determines the loan history as “present” (Yes in S602), and is not so. In this case, it may be determined as “none” (No in S602).

  The provisional score determination unit 12 selects model 1 if Yes in S601 and No in S602 (S603), selects Model 2 if Yes in S601 and Yes in S602 (S604), and if No in S601, the model. 3 is selected (S605).

  In this way, changing the model for calculating the DF rate according to the customer's financial institution internal information (deposit information, past loan information, etc.) examines the relationship between the customer who applied for the loan and the financial institution. In other words, this is to reflect the score (score) for calculating the DF rate.

  For example, a customer corresponding to Model 1 is an initial loan that is a good deal for the financial institution, and although it is expected to lead to a future borrowing, there is no information on the ability to repay the loan. Compared with the customer of the model 1, the customer corresponding to the model 2 has a lot of information for determining the score based on the loan repayment ability because the financial institution has the past loan actual data. Further, since the customer corresponding to the model 3 is a so-called first-looker for the financial institution, there is little information for score determination. In this way, it is possible to perform the loan screening process according to the relationship between the customer and the financial institution so far, instead of screening all the customers who have applied for the loan based on a uniform standard.

  Although FIG. 6 shows that the model for calculating the DF rate is selected based on the information on the deposit transaction history and the past loan history of the internal information of the financial institution, the internal structure of the financial institution used when selecting the model is shown. The information is not limited to this example. It is sufficient if the difference in the model reflects the relationship between the customer and the financial institution. For example, the model may be selected using the deposit amount of the financial institution internal information. In this case, if the deposit amount is small, the relationship between the customer and the financial institution is considered to be thin (in other words, the customer's contribution to the financial institution is small), and if the deposit amount is large, the relationship between the customer and the financial institution. It is considered that there is a strong nature (in other words, the degree of contribution of customers to financial institutions is large). Therefore, multiple models may be designed so that the score calculated using the model selected when the deposit amount is small is smaller than the score obtained with the model selected when the deposit amount is large . In addition, for example, the model may be selected based on information such as the presence / absence of a time deposit account and the presence / absence of a salary transfer account in the internal information of the financial institution. In this case, a customer who has a fixed deposit account or a customer who has a salary transfer account is considered to have a relatively strong relationship with the financial institution, and therefore the relationship may be reflected in the model.

  Next, an example of calculation of the DF rate based on each model will be described with reference to FIGS. FIG. 7 is a flowchart of the DF rate calculation process performed by the provisional score determination unit 12.

  In FIG. 7, first, the provisional score determination unit 12 reads the scoring tables 17a, 17b, or 17c corresponding to the models 1 to 3 stored in advance in the auxiliary storage unit of the loan examination support device 10 in S701. That is, the provisional score determination unit 12 reads the scoring table 17a corresponding to the model 1 when the model 1 is selected in the process S503, and the scoring table 17a corresponding to the model 2 when the model 2 is selected. When 17b is read and model 3 is selected, control is performed so that scoring tables 17a and 17c corresponding to model 3 are read. The reason why the scoring table 17a is also read in the models 2 and 3 is that the value of the total score Z calculated based on the scoring table 17a is used in the scoring tables 17b and 17c, respectively.

  FIG. 8 shows an example of the scoring table 17a corresponding to the model 1. 9 shows an example of the scoring table 17b corresponding to the model 2, and FIG. 10 shows an example of the scoring table 17c corresponding to the model 3.

  The scoring table 17a corresponding to the model 1 shown in FIG. 8 includes, as one embodiment, “deposit history”, “account transaction history”, “work industry”, “employment” among customer application information and financial institution internal information. It has information on each item of “form”, “position”, “months of service”, “residence classification”, and “age at the time of financing”, and is used to calculate the total score Z for the model 1. The scoring table 17b corresponding to the model 2 shown in FIG. 9 has information on “past loan history” among the financial institution internal information, and is used to calculate the total score Z for the model 2. The scoring table 17c corresponding to the model 3 shown in FIG. 10 is used to calculate the total score Z for the model 3 based on the total score Z calculated using the scoring table 17a. Note that the item of “personal credit information” of a customer acquired and accumulated in the past from an external credit agency or the like may be included in the scoring table 17. For example, when information with personal credit information “x” exists in the financial institution internal information, the score may be negative (for example, −10 to −50).

  Next, in S702, the provisional score determination unit 12 extracts information necessary for calculating the total score Z from the read scoring tables 17a to 17c from the application information and the internal information of the financial institution. For example, in order to calculate the total score Z for model 1, “deposit history”, “account transaction history”, “workplace industry”, “employment form”, “job title” from customer application information and financial institution internal information , Information on each item of “month of service”, “residence classification”, and “age at the time of financing” is extracted.

  In S703, the provisional score determination unit 12 determines which condition in each item the extracted information corresponds to, and calculates the total score Z by adding the corresponding score z. Specifically, the provisional score determination unit 12, as one embodiment, calculates an average of deposits for a plurality of months before a loan application date (or a loan execution date or a loan examination date) based on information related to “deposit history” of financial institution internal information. The amount is calculated, and the value of the rating z of the corresponding condition in the item “deposit history” is read. In addition, as an embodiment, the provisional score determination unit 12 determines “whether or not there is salary transfer (payment) or pension receipt” based on information related to “account transaction history” in the internal information of the customer's financial institution, and the item “ In the “account transaction history”, the value of the score z of the corresponding condition is read. Note that the number of months varies depending on the analysis result of the internal information of the financial institution when the scoring table 17 is created. For example, 2 to 24 months, 2 to 18 months, 3 to 12 months, or 3 to 3 months It is possible to employ various months such as 6 months.

  In this way, the provisional score determination unit 12 includes the “deposit history”, “account transaction history”, “work industry”, “working style”, “position”, “month of service”, “residential classification”, and The value of the corresponding score z is read for all items of “age at the time of financing”, and finally, the total score Z is calculated by adding all the read scores z.

  Further, the provisional score determination unit 12 calculates the total score Z for the models 2 and 3 in the same manner as the model 1 described above. Specifically, as illustrated in FIG. 9, the provisional score determination unit 12 uses the scoring table 17b of the model 2 and, as one embodiment, based on information related to “past loan history”, The presence or absence is determined, and the corresponding score z is read as the score z of the item “past loan history”. Then, the provisional score determination unit 12 sums this value and a value obtained by multiplying the total score calculated by using the scoring table 17a of the model 1 by a predetermined constant (0.8 as an example) to obtain the model 2 The total score Z is calculated. Also, as shown in FIG. 10, the provisional score determination unit 12 uses the scoring table 17c of the model 3, and as one embodiment, relates to each item of “deposit history” and “account transaction history” of the scoring table 17a. The total score Z of the model 3 is calculated by calculating the total score Z according to the scoring table 17a after setting all the scores to 0, and multiplying the total score Z by a predetermined constant (1.2 as an example).

In S704, the provisional score determination unit 12 calculates the DF rate (%) of the customer in the application loan based on the following formula from the total score Z calculated in S703, and returns the process to S505.

Where DFR is the DF rate and Z is the total score for the model selected in S503. H ₀ is a constant that varies depending on the models 1 to 3 and the loan type (target loan, free loan, card loan, etc.), and γ is a constant that varies depending on the models 1 to 3. Although not limited, examples of H ₀ and γ are shown in Tables 1 and 2 below. The values of the constants H ₀ and γ are constants obtained recursively by analyzing the internal information of the financial institution (Poisson process or the like), as in the creation of a scoring table described later.

  Next, in S505, the provisional score determination unit 12 determines a rank based on the calculated DF rate. Although not limited, FIG. 11 shows an example of a table showing the relationship between rank and DF rate. A table indicating the relationship between the rank and the DF rate is stored in advance in the loan examination support apparatus 10, and the provisional score determination unit 12 reads the table and determines the rank based on the DF rate calculated in S704.

  In the example of FIG. 11, rank R01 is the highest rank and R11 is the lowest rank. For example, R01 to R02 have very low credit risk, R03 to R04 have low credit risk, R05 has average credit risk, R06 to R07 have high credit risk, R08 to R09 have very high credit risk, R10 ˜R11 may be classified as an infeasible rank of financing. In this example, the rank is classified into 11 stages of R01 to R11, but the present invention is not limited to this. The rank may be classified into at least 2 to 3 levels. According to the example of FIG. 11, if the DF rate calculated in S704 is 0.7%, the rank is determined as R04.

<Create scoring table>
Here, creation of the scoring table 17 and calculation of the DF rate will be described. The scoring table 17 is created by analyzing internal information of financial institutions (a technique such as a Poisson process), and is stored in advance in the loan examination support device 10.

  With reference to the example of the scoring table 17a shown in FIG. 8, how to assign a score for each condition in each item (deposit history, account transaction history, etc.) of the scoring table will be described. For example, with regard to the item “deposit history”, internal information of financial institutions is analyzed, and the average number of deposits over the last several months (for example, 3 to June may be more than 1,000,000 yen) and the number of people who caused defaults Based on this, the DF rate of the person who meets this condition is calculated (0.1% in the DF rate column in the scoring table 17a in FIG. 8). Also, based on the number of people whose average multi-month deposit was more than 500,000 yen to 1,000,000 yen and the number of people who caused the default among them, the DF rate of those who meet this condition is calculated (0.3%). In this way, the DF rate is also calculated for other conditions in the item “deposit history”. Then, the DF rate is similarly calculated for each condition of other items (account transaction history and the like).

  Thereafter, a maximum score in each item (deposit history, account transaction history, etc.) is allocated so that the total maximum score of each item in the scoring table 17a is about 100. The maximum score is assigned to each item (deposit history, account transaction history, etc.) empirically (regressively). For example, in the example of the scoring table 17a of FIG. 8, the maximum score assigned to the item “deposit history” is 40, and the maximum score assigned to the item “account transaction history” is 10.

In addition, in assigning a score to each condition in each item (deposit history, account transaction history, etc.), the score for the condition showing the maximum DF rate in each item is 0, and the score for the condition showing the minimum DF rate is A linear expression with a maximum score assigned and a straight line between the two:
Y = {maximum score / (maximum DF rate−minimum DF rate)} × (maximum DF rate−X)
The value of Y obtained by substituting the DF rate of each condition into X is assigned to each condition as a score. In this way, a score can be assigned to each condition of each item.

  As described above, the scoring table 17 is created in advance by analyzing the internal information of the financial institution, and is stored in advance in the loan examination support device 10 (or the database 20). Since the internal information of the financial institution changes day by day, the scoring table 17 may be updated every predetermined period (for example, but not limited to several weeks to several months).

<Proper credit judgment processing>
Next, details of the proper credit determination process (S304) will be described with reference to FIG. In the proper credit determination process, a determination is made as to whether or not credit for proper financing is available.

  In FIG. 12, first, in S1201, the loan examination support device 10 transmits inquiry data based on customer application information or financial institution internal information to the foreign trust determination device 13, and the foreign trust determination device 13 responds accordingly. Access one or more external credit terminal 40 to obtain personal credit information about the customer. The acquired personal credit information includes, as one embodiment, comprehensive determination processing results, comprehensive determination processing details, care mark information, contract status, contract summary information, inquiry history information, and payment history information.

  In step S <b> 1202, the foreign communication determination device 13 performs a foreign communication determination process based on the acquired personal credit information. In the foreign credit determination process, the creditworthiness of the customer is evaluated in a plurality of stages, for example, two stages (◯ or X) or three stages (◯, Δ, or X) based on personal credit information provided from an external credit agency.

  For example, if the foreign credit determination device 13 determines that there is a serious financial accident information such as “transfer” in the personal credit information provided by the external credit agency, the credit rating of the customer is determined as “×” (severe If there is no such serious financial accident, it is determined as “◯” (no problem) or “△” (examiner check required). The foreign communication determination device 13 transmits the result of the foreign communication determination processing (such as ◯, Δ, or ×) to the loan examination support device 10.

  In addition, a foreign credit determination unit is provided in the loan examination support device 10, and the foreign trust determination unit receives data obtained by analyzing and collecting personal credit information from the foreign trust determination device 13. May be performed. It should be noted that system processing according to the determination result of the foreign communication determination processing in S1202 (for example, system processing for a case where the foreign communication determination is not OK) is not the subject of the present invention, and therefore detailed descriptions thereof will be omitted. In the subsequent flow, the description will be continued as a flow for the matter processing that has successfully passed the foreign communication determination processing.

  In S1203, the score determination unit 14 selects a model to be used for calculating the DF rate based on customer application information and financial institution internal information. In step S1204, the score determination unit 14 calculates the DF rate of the customer in the application loan based on the selected model. In step S1205, the score determination unit 14 determines a rank based on the calculated DF rate. Since the processing of S1203 to S1205 is the same as the processing of S503 to S505 described above, description thereof will be omitted.

  In S1206, the loan amount determination unit 15 extracts (1) the desired borrowing amount related to the application for the loan and (2) the borrowing amount related to the existing loan based on the application information, and based on the internal information of the financial institution (3) The existing borrowing amount is extracted, and the information on the existing borrowing amount in another financial institution or the like acquired by the foreign credit determination device 13 from the terminal 40 of the external credit institution is extracted.

  In S1207, the loan amount determination unit 15 calculates the total amount of the borrowing amount of (1) to (4) except for the overlapping portion of the borrowing amount of (2) to (4), and calculates the total borrowing amount, the repayment period, Calculate the “repayment ratio” (%) from the formula “total annual repayment amount ÷ annual income” based on the annual income, and calculate the “annual income ratio” (%) based on the formula “total borrowing ÷ annual income” To do.

In S1208, the loan amount determination unit 15 performs a loan amount determination process (◯, Δ, or ×) based on the determination criteria shown in Table 3 below. The determination items (repayment ratio and annual income ratio) and their numerical criteria shown in Table 3 are examples, and the present invention is not limited thereto.

  In S1209, the credit determination unit 16 determines whether or not proper credit is possible based on the results of the external communication determination process in S1202, the rank determination in S1205, and the loan amount determination process in S1208. The proper credit determination process is performed based on the proper credit determination table shown in FIG. For example, if the result of the foreign credit determination process is “Δ”, the credit determination unit 16 sets the result of the proper credit determination to “Δ” (need to be checked by the examiner) regardless of the rank value. Further, if the result of the foreign credit determination process is “O”, the rank is R01, and the result of the loan amount determination process is “O” or “Δ”, the credit determination unit 16 indicates the result of the proper credit determination process. “O” (no problem). When the result of the foreign credit determination process is “◯”, the rank is R08, and the result of the loan amount determination process is “Δ” or “×”, the credit determination unit 16 sets the result of the proper credit determination process as “ X ”(severe problem). The proper credit determination table shown in FIG. 13 is an example of proper credit determination criteria, and the present invention is not limited to this.

  If the proper credit determination process of S1210 results in OK (◯ or Δ) (that is, when it is determined that proper financing is possible), the result is sent from the loan screening support apparatus 10 to the screening terminal 30. The person in charge of examination and the person in charge of examination make a final decision as to whether or not to accept the loan (S305 to S306), and the customer is notified of the result.

  On the other hand, if the proper credit determination process in S1210 is not OK (×) (when it is determined that proper loan is not possible), the loan examination support apparatus 10 applies for the customer loan application to the terminal 50 of the external guarantee company in S1211. Send inquiry data such as such information and make a loan guarantee request to the guarantee company. Instead of directly transmitting the inquiry data from the loan examination support device 10 to the terminal 50, a document relating to the guarantee request may be sent to the guarantee company by FAX or mail by the person in charge of the financial institution. If the guarantee company guarantees, the financial institution will pay the guarantee fee to the guarantee company, but if it accepts the loan application and the guarantee company does not guarantee it, the financial institution will And the result is communicated to the customer. This is the same after the processing of S508 described above.

  As described above, FIG. 14 shows an example of the correspondence relationship between ranks R01 to R10, estimated annual DF rate, guarantee fee, credit cost, and annual expected sheath obtained by analyzing internal information of financial institutions. It is a table | surface which shows. As can be seen from this table, the correspondence between the rank (or DF rate) derived by the loan examination support device 10 according to this embodiment and the expected sheath (expected profit) is clear, and the expected sheath is a predetermined value (for example, Proper loans (percentage) that do not request a guarantee from the guarantee company can be increased as long as they do not become negative. As a result, the guarantee fee paid to the guarantee company for the increase in proper financing is reduced.

[Second Embodiment]
The loan examination support apparatus according to the second embodiment of the present invention can be used in loan examination for a new loan application, and is also used in the case of an ongoing credit for a case where a loan has already been executed (stock bond). be able to. Note that the loan examination support apparatus of the present embodiment is basically the same as the loan examination support apparatus 10 of the first embodiment, and a description thereof will be omitted. In the present embodiment, the credit determination unit 16 of the loan examination support device 10 finally determines whether or not the guaranteed loan being executed by the financial institution can be changed to the proper loan. In the present embodiment, the credit determination unit 16 determines whether the credit of the customer is improved or deteriorated when the customer is borrowing with an unsecured loan, and the credit is improved or deteriorated. In other words, it may be determined that the credit line of the customer's unsecured loan should be expanded or contracted, respectively.

  The loan examination support apparatus according to the second embodiment determines whether or not proper credit is possible for each of the cases where the existing loan is being executed, in the same way as in the first embodiment, in the case of mid-term credit. For cases where proper credit is possible, it can be changed to proper financing at the discretion of the examiner and payment of the guarantee fee to the guarantee company can be stopped. As a result, the guarantee fee is reduced and the profit of the financial institution is improved.

  For example, FIG. 15 shows ranks R01 to R10, an estimated annual DF rate, a guarantee fee, a credit cost, and an annual expected sheath obtained by analyzing financial institution internal information regarding an existing loan execution case of the financial institution. It is a table | surface which shows these relationships. FIG. 15A shows the case of a target loan, and FIG. 15B shows the case of a free loan. In FIG. 15, “guarantee fee” is a cost (cost) generated by requesting a guarantee from a guarantee company, and “credit cost” is caused by a financial institution taking a risk when the guarantee of the guarantee company is stopped. It is a potential expense (cost), and “expected sheath” indicates an expected profit generated when the guarantee fee is set to zero. For example, in the case of FIG. 15 (a), the expected pod is positive up to ranks R01 to R05, so the existing loan case corresponding to ranks R01 to R05 is changed to proper lending, so that guarantee The fee is reduced, and the financial institution can obtain profits (in this example, the total expected sheath of R01 to R05 is approximately 25,800,000 yen / year).

  Further, when a customer borrows an unsecured loan from a financial institution, the score determination unit 14 determines the DF rate and rank of the customer at the time of credit on the way. When the determined rank is improved to a rank higher than a predetermined rank number (for example, R01) or deteriorated to a rank lower than a predetermined rank number (for example, R05) from the rank at the time of loan examination (for example, R03) The credit determination unit 16 determines that the credit limit of the unsecured loan of the customer should be expanded or contracted, and outputs the result to the examination terminal 30. In response to this result, the examiner decides whether to expand or contract the credit line of the customer whose rank is improving or deteriorating. This makes it possible to increase the profit of the financial institution or reduce the risk during the credit process.

[Third Embodiment]
The loan examination support apparatus according to the third embodiment of the present invention is based on information about a customer (not limited to whether or not a loan is received from the financial institution) stored in the financial institution, and a loan solicitation target. Can be extracted and used for future marketing. Note that the loan examination support apparatus of the present embodiment is basically the same as the loan examination support apparatus 10 of the first and second embodiments, and a description thereof will be omitted. The loan examination support device according to the present embodiment is also referred to as a marketing support device.

  In the present embodiment, the score determination unit 14 of the loan examination support device 10 selects a model, calculates a DF rate, and determines a rank based on internal information of a financial institution, and extracts a target customer (in this case The extraction rank can be set arbitrarily).

  The loan examination support apparatus according to the third embodiment extracts information of a plurality of customers from the internal information of the financial institution and ranks each customer based on the internal information of the financial institution regarding each customer. At the time of ranking, the DF rate is calculated using the scoring table (in FIG. 8 to FIG. 10 in one example), and ranking is performed therefrom (in FIG. 11 in one example). Then, the loan examination support apparatus according to the present embodiment associates customer information (name, address, telephone number, etc.) included in the internal information of the financial institution with the determined rank, and outputs it to a terminal in the financial institution. Customers with a high rank (for example, R01 to R03) have a low DF rate, so there is little risk of credit loss. Therefore, by using the loan examination support apparatus according to the present embodiment, it becomes possible for a financial institution to take a strategy of proactively recommending loans to such target customers.

  Further, the financial institution may extract or output a change in the rank of the customer over time by using the loan examination support device, and may use it for marketing such as solicitation of loans. For example, in the loan examination support apparatus according to the third embodiment, the rank of the customer at a certain time (for example, R07) is a rank that is equal to or higher than a predetermined rank value (for example, R03) at a later time (for example, two to three years later). Customer information such as (for example, R02) can be extracted. Further, the loan examination support device according to the present embodiment determines the rank of the customer at a certain time interval (for example, every three months), and the average rank of the previous ranks at a certain point in time becomes equal to or higher than a predetermined rank. Customer information can also be extracted. With such extracted information, customers whose rank has improved over time are considered to be customers who have steadily gained creditworthiness, so such customers are considered to have a low risk of bad debt for a while, Can actively promote solicitation of loans such as target loans and proposals to expand credit lines (the success rate of such positive measures is expected to be high). Conversely, customers whose rank has declined over time are considered to be experiencing a risk of bad debts due to a decline in creditworthiness, so they should be used as judgment materials for adopting a reluctance to not solicit loans. be able to.

  Soliciting loans to customers can also be automated. For example, a loan examination support apparatus and a server that manages ATM are communicably connected, and the loan examination support apparatus transmits information about a customer having a high rank to the server. When the target customer operates the ATM, the server displays the solicitation information (loan type, interest rate, possible loan amount, etc.) on the ATM screen or issued statement. Also good.

[Fourth Embodiment]
The loan examination support apparatus according to the fourth embodiment of the present invention performs examination for a plurality of types of loans at once (or sequentially) with one loan application. Note that the loan examination support device of the present embodiment is the same as the loan examination support device 10 of the first to third embodiments, and a description thereof will be omitted.

  In the fourth embodiment, one type of loan application is used to examine multiple types of loans. However, the calculation of the DF rate, the assignment of ranks, and the determination of the loan amount are performed for each type of loan, and credit determination is performed. The result of the process and / or the foreign currency determination process can be used in common for all loan applications. As shown in Table 1 above, the constants used in the formula for calculating the DF rate may be different for each type of loan, so the rank may be different for each type of loan. Therefore, the proper credit judgment results for all types of loans do not match.

  As an example of the operational effect of the loan examination support apparatus according to the fourth embodiment, simultaneous examination of a target loan and a card loan based on one loan application can be cited. Such simultaneous examination (or sequential examination) has the effect of being able to respond flexibly without losing the opportunity to provide services for the applicant's funding needs.

  In addition to the benefits of reducing the input procedures and screening work based on a single loan application, guarantors and joint debtors can also be granted for each loan in the application, allowing flexibility in credit judgment There is also an effect that the property is improved.

  Further, as an implementation example for realizing the loan examination support apparatus according to the fourth embodiment, it is realized by performing unified management for uniquely managing the applicant with the “applicant identification number” already described. In this case, more specifically, information at the time of application and information after execution of the loan are linked on the data table, and storage from the time of loan application to the end of the loan is stored and managed in the database. When such management operation is performed, a person who has not been connected with the loan contract can be uniquely identified by the applicant identification number, and can be used for the above-described marketing, thereby increasing the synergistic effect.

  As an example of the data format of the applicant identification number in the present embodiment, a format such as a unique serial number of 8 digits can be adopted. When applying for a loan, the applicant identification number is newly assigned in the system if the customer is a new customer, and the customer's applicant number is used for the existing customer if the customer is an existing customer.

[Fifth Embodiment]
In the loan examination support apparatus according to the fifth embodiment of the present invention, in addition to adopting the applicant identification number (fourth embodiment), a serial number related to application information in order to perform history management of application information including application input information. Management is adopted. In such an embodiment, for example, when an application change process is made after a proper loan (to be) determination or a guaranteed loan (to be) determination, the contents when the determination is made before the change Since there is a possibility of difference, it is used for processing for confirming consistency with application information at the time of determination before change.

As the management number related to the application, the management numbers shown in the table below can be adopted.

  The above embodiment is changed according to the structure of the apparatus to which the present invention is applied or various conditions. Further, the present invention is not limited to the above-described embodiment specifically described.

10 Loan Examination Support Device 11 Outline Reference Determining Unit 12 Temporary Score Determining Unit 13
14 Score Determination Unit 15 Loan Amount Determination Unit 16 Credit Determination Unit 17 Scoring Table 20 Database 30 Screening Terminal 35 Office Terminal 40 External Credit Institution Terminal 50 Guarantor Company Terminal 100 Loan Screening Support System

Claims (17)

A score determination unit that calculates the default rate of the customer and determines the rank;
A loan examination support device comprising a credit judgment unit for judging whether proper loan for an application loan is possible,
The score determination unit
A plurality of models representing relationships between loan application information and / or internal financial institution information about customers and scores used to calculate customer default rates, each of which has a different relationship among the models To select at least one model based on internal financial institution information about the customer,
Based on the loan application information and / or financial institution internal information about the customer, the default rate of the customer is calculated from the selected model,
A rank is determined based on the calculated default rate,
The credit judgment unit
Determining whether proper financing for the application loan is possible based on at least the rank;
A loan examination support device characterized by that.   The model selected by the score determination unit when there is deposit information in the internal information of the financial institution related to the customer is the model selected when the score is assigned to the customer's deposit information and there is no deposit information The loan examination support apparatus according to claim 1, wherein the rating is not assigned to customer deposit information.   The model selected by the score determination unit when there is past loan information in the financial institution internal information about the customer is selected when the score is assigned to the customer's past loan information and there is no past loan information The loan examination support apparatus according to claim 1, wherein the score is not assigned to the customer's past loan information.   Loan application information and / or internal information about financial institutions related to customers includes: whether there is a deposit account, deposit account opening date, deposit history, account transaction history, deposit balance, salary or pension transfer, whether there is a utility charge deduction , Customer's annual income, family structure, workplace industry, employment type, job title, length of service, residence classification, housing type, loan age, past loan history, customer applied to the financial institution in the past The loan type, the number of applications, the number of loans, the loan amount, the unpaid amount, the repayment status, the accident information related to the repayment of the loan, and at least one of personal credit information are included. 4. The loan examination support device according to any one of items 3. A foreign credit judgment unit that judges whether personal credit information about customers provided by an external credit agency contains information on a serious financial accident;
A loan amount determination unit that calculates the annual income ratio and repayment ratio of the customer based on the loan application information and / or the internal information of the financial institution relating to the customer;
The credit determination unit determines whether proper loan for the application loan is possible based on the rank, the determination result by the foreign credit determination unit, and the annual income ratio and the repayment ratio calculated by the loan amount determination unit. The loan examination support apparatus according to any one of claims 1 to 4, wherein A score determination unit that calculates the default rate of the customer and determines the rank;
A loan examination support device comprising a credit judgment unit for judging whether or not a guaranteed loan being executed by a financial institution can be changed to a proper loan,
The score determination unit
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Select at least one model based on information,
Based on the financial institution internal information about the customer, calculate the default rate of the customer from the selected model,
A rank is determined based on the calculated default rate,
The credit judgment unit
Determining whether the guaranteed loan can be changed to a proper loan based on at least the rank;
A loan examination support device characterized by that. A score determination unit that calculates the default rate of the customer and determines the rank;
A loan examination support device comprising a credit judgment unit that judges whether or not the credit limit of a customer should be reduced in a credit in progress of a loan being executed by a financial institution,
The score determination unit
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Select at least one model based on information,
Based on the financial institution internal information about the customer, calculate the default rate of the customer from the selected model,
A rank is determined based on the calculated default rate,
The credit judgment unit
The rank of the customer determined by the score determination unit at the time of entrusting credit has improved to a rank higher than a predetermined rank number compared to the rank of the customer determined by the score determination unit at the time of loan examination If it is determined that the credit line of the customer should be expanded, or if it has deteriorated to a rank lower than a predetermined number of ranks, it is determined that the credit line of the customer should be reduced,
A loan examination support device characterized by that.   A loan examination support system including the loan examination support apparatus according to claim 1. A score determination process that calculates the default rate of the customer and determines the rank;
A program for causing a computer to execute a credit judgment process for judging whether or not proper financing for an application loan is possible,
The score determination process includes
A plurality of models representing relationships between loan application information and / or internal financial institution information about customers and scores used to calculate customer default rates, each of which has a different relationship among the models To select at least one model based on internal financial institution information about the customer,
Processing for calculating a default rate of a customer from the selected model based on loan application information and / or financial institution internal information about the customer, and processing for determining a rank based on the calculated default rate,
Including
The credit determination process includes
A program comprising: determining whether or not a proper loan for the application loan is possible based on at least the rank.   The model selected by the score determination process when the deposit information is included in the internal information of the financial institution related to the customer is the model selected when the score is assigned to the deposit information of the customer and there is no deposit information 10. The program according to claim 9, wherein the score is not assigned to customer deposit information.   The model selected by the score determination process when there is past loan information in the internal information of the financial institution related to the customer is selected when the score is assigned to the customer's past loan information and there is no past loan information 11. The program according to claim 9, wherein the rating is not assigned to a customer's past loan information.   Loan application information and / or internal information about financial institutions related to customers includes: whether there is a deposit account, deposit account opening date, deposit history, account transaction history, deposit balance, salary or pension transfer, whether there is a utility charge deduction , Customer's annual income, family structure, workplace industry, employment type, job title, length of service, residence classification, housing type, loan age, past loan history, customer applied to the financial institution in the past The loan type, the number of applications, the number of loans, the loan amount, the unpaid amount, the repayment status, the accident information related to the repayment of the loan, and at least one of personal credit information are included. 12. The program according to any one of 11 above. Foreign credit judgment processing to determine whether personal credit information about customers provided by external credit agencies contains information on serious financial accidents,
A loan amount determination process for calculating a customer's annual income ratio and repayment ratio based on loan application information and / or internal financial institution information about the customer,
The credit determination process determines whether proper loan for the application loan is possible based on the rank, the determination result by the foreign credit determination process, and the annual income ratio and the repayment ratio calculated by the loan amount determination process. The program according to any one of claims 9 to 12. A score determination process that calculates the default rate of the customer and determines the rank;
A program for causing a computer to execute a credit determination process for determining whether a guaranteed loan being executed by a financial institution can be changed to a proper loan,
The score determination process includes
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Selecting at least one model based on information,
A process for calculating a default rate of a customer from the selected model, based on internal information about the financial institution relating to the customer, and a process for determining a rank based on the calculated default rate;
Including
The credit determination process includes
A process of determining whether the guaranteed loan can be changed to a proper loan based on at least the rank;
The program characterized by including. A score determination process that calculates the default rate of the customer and determines the rank;
A program for causing a computer to execute a credit determination process for determining whether or not to reduce a credit line of a customer in a credit in progress of a loan being executed by a financial institution,
The score determination process includes
A plurality of models representing a relationship between financial institution internal information relating to a customer and a score used for calculating a default rate of the customer, and each of the models having different relationships from each other, the financial institution relating to the customer Selecting at least one model based on information,
A process of calculating a default rate of the customer from the selected model based on internal information about the financial institution relating to the customer, and a process of determining a rank based on the calculated default rate,
The credit determination process includes
The rank of the customer determined by the score determination unit at the time of entrusting credit has improved to a rank higher than a predetermined rank number compared to the rank of the customer determined by the score determination unit at the time of loan examination The credit line of the customer is determined to be expanded, or the process of determining that the credit line of the customer should be reduced if the customer's credit limit is deteriorated to a rank lower than a predetermined number of ranks. Program.   The processes for calculating the default rate and determining the rank are performed for each of a plurality of types of loans, and the credit determination process is performed for each of the plurality of types of loans based on the loan application information. The apparatus according to claim 1. The default rate calculation process and the rank determination process are performed for each of a plurality of types of loans, and the credit determination process is performed for each of the plurality of types of loans based on the loan application information. The program according to claim 9.