JP2014026322A - Examination model creation device, computer program, and examination model creation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily set a proper interest rate to profit of a financial institution without increasing a burden of the financial institution by creating an examination model with a change in default and a change in prepayment taken into consideration.SOLUTION: An examination model creation device 1 includes a default model creation part 10 for creating a default model for a loan applicant, a prepayment model creation part 12 for creating a prepayment model for an applicant, a fund raising cost calculation part 18 for calculating fund raising cost corresponding to the change in prepayment by using the prepayment model, and an interest rate spread calculation part 20 for calculating an interest rate spread corresponding to the change in default and the change in prepayment by using the default model, the prepayment model, and the fund raising cost.

Description

  Embodiments described herein relate generally to an examination model creation apparatus, a computer program, and an examination model creation method for creating an examination model used for examination of credit quality of a loan applicant.

  In financial institutions, applicants are screened when deciding whether to accept a mortgage application. In this examination, an examination system that calculates the creditworthiness of the applicant is used to eliminate the subjective judgment of the person in charge of examination and shorten the examination period.

  For example, when the examiner gives the examination system information necessary for the examination (information regarding the attributes of the applicant, loan conditions, collateral information) to the examination system, the examination system displays a credit index ( A credit score and a repayment impossible probability (hereinafter referred to as “PD (Probability of Default)”)) are calculated. The examiner determines whether to accept the application for the mortgage based on the credit index calculated by the examination system.

  Conventional screening systems use credit models (eg, from the loan date to the loan date) using a review model (eg, logistic regression model or discriminant tree model) to relatively assess the credit risk of multiple obligors. Calculate the PD of the period. The examiner determines the applicable interest rate to be applied to the mortgage in consideration of this PD and loan costs (for example, funding costs and expenses).

  A conventional examination model is created using samples of loans that have been loaned in the past few years. Specifically, the conventional examination model expresses the relationship between the credit, the debtor, and the PD from the credit data, the debtor data, and the bad debt data. However, the conventional examination model has the following problems.

  First, if the sample is composed mainly of receivables with a short elapsed time since the start of lending (hereinafter referred to as “short-term accrued receivables”), the problem is that the sample PD deviates from the population PD. is there.

  In order to calculate the PD for each bond, the PD obtained from the population is necessary. However, in practice, the PD cannot be obtained from the population. Therefore, in practice, PD estimated from a sample is used. However, for short-term accrued loans (for example, mortgages executed within 5 years), you can obtain short-term PD when the elapsed time from the start of the loan is short, but when the elapsed time from the start of the loan is long Long-term PD cannot be obtained. This causes the sample PD to deviate from the population PD.

  Secondly, there is a problem that the risk of bad debt cannot be fully evaluated.

  In the mortgage, a change with time (hereinafter referred to as “change over time”) (for example, change in PD, change in repayment amount due to debtor's prepayment, change in repayment amount due to change in interest rate) occurs. This change over time makes it difficult to assess the risk of credit loss.

  Third, there is a problem that the cost for the prepayment option is not considered in the calculation of the loan cost.

  The funding rate of a financial institution is calculated in consideration of prepayments in addition to scheduled repayments. If mortgage prepayment depends on the obligor's will, it can be understood that the mortgage has a prepayment option. This prepayment option can be a risk for financial institutions and should be considered in calculating this loan cost. However, with the conventional screening model, it is difficult to calculate the loan cost considering the prepayment option.

JP 2002-197268 A

  The problem to be solved by the present invention is to create an examination model that takes into account changes in defaults and changes in prepayment, thereby facilitating an appropriate interest rate for financial institution revenue without increasing the burden on the financial institution. Is to calculate.

According to the first aspect of the present invention,
A default model creation section for creating a default model for loan applicants;
A prepayment model creation unit for creating a prepayment model of the applicant;
Using the prepayment model, a funding cost calculation unit that calculates a funding cost according to a change in the prepayment,
An interest rate spread calculation unit for calculating an interest rate spread according to a change in default and a change in the payment using the default model, the prepayment model, and the funding cost;
An examination model creation device characterized by comprising:
Is provided.

According to a second aspect of the invention,
On the computer,
Creating a default model for loan applicants;
Creating a prepayment model for the applicant;
Using the prepayment model to calculate funding costs in response to changes in the prepayment;
Using the default model, the prepayment model, and the funding cost to calculate an interest rate spread according to a change in default and a change in the prepayment;
A computer program that runs,
Is provided.

According to a third aspect of the invention,
Computer
Creating a default model for loan applicants;
Creating a prepayment model for the applicant;
Using the prepayment model to calculate funding costs in response to changes in the prepayment;
Using the default model, the prepayment model, and the funding cost to calculate an interest rate spread according to a change in default and a change in the prepayment;
To create a review model,
Is provided.

  According to the present invention, it is possible to easily calculate an interest rate appropriate for the profit of a financial institution without increasing the burden on the financial institution.

The block diagram which shows the structure of the examination system of this embodiment. Explanatory drawing of the data which the examination model creation apparatus 1 of this embodiment handles. The figure which shows the data structure of the obligor attribute data 31 of this embodiment. The figure which shows the data structure of the past loan data 35 of this embodiment. The flowchart which shows the procedure of the examination model creation process of this embodiment. The flowchart which shows the procedure of calculation (S1-18 of FIG. 5) of the funding cost of this embodiment. The flowchart which shows the procedure of calculation (S1-20 of FIG. 5) of the interest rate spread of this embodiment.

  The present embodiment will be described with reference to the drawings.

  A configuration of the examination model creation device 1 of the present embodiment will be described. FIG. 1 is a block diagram showing the configuration of the examination system of this embodiment.

  As shown in FIG. 1, the screening system includes a screening model creation device 1, a database 3, an input device 5, an output device 7, and a communication device 9.

  The screening model creation device 1 is a device that creates a screening model used for screening a loan (for example, mortgage) applicant. The examination model creation device 1 includes a default model creation unit 10, a prepayment model creation unit 12, a yield curve creation unit 14, an interest rate model parameter estimation unit 16, a funding cost calculation unit 18, and an interest rate spread calculation unit 20. (Each part of the examination model creation device 1 will be described later). For example, the examination model creation device 1 is a computer processor that implements each part of the examination model creation device 1 by executing a computer program.

  The database 3 is a device that stores data referred to by the examination model creation device 1 (that is, data used for creation of the examination model).

  The input device 5 is a device that receives an instruction input to the examination model creation device 1. For example, the input device 5 is at least one of a keyboard and a mouse. The person in charge of examination can give an instruction to the examination model creation apparatus 1 via the input device 5.

  The output device 7 is a device that outputs the processing result of the examination model creation device 1. For example, the output device 7 is at least one of a printer and a display. The person in charge of examination can confirm the processing result of the examination model creation apparatus 1 via the output device 7.

  The communication device 9 is a device that controls communication between the examination model creation device 1 and a device outside the examination system. For example, the communication device 9 is an internet or intranet communication interface. The examination model creation device 1 can receive an instruction to the examination model creation device 1 via the communication device 9 and output the processing result of the examination model creation device 1.

  Data referred to by the examination model creation device 1 of the present embodiment will be described. FIG. 2 is an explanatory diagram of data handled by the examination model creation device 1 of the present embodiment. FIG. 3 is a diagram illustrating a data structure of the obligor attribute data 31 of the present embodiment. FIG. 4 is a diagram showing a data structure of the past loan data 35 of the present embodiment.

  As shown in FIG. 2, the database 3 stores debtor attribute data 31, loan repayment history data 32, market data 33, loan application data 34, and past loan data 35.

  The obligor attribute data 31 is obligor attribute information (for example, obligor number identifying the obligor, residence, gender, occupation, date of birth, etc.) of the loan executed in the past (hereinafter referred to as “past loan”). And annual income etc.) (see FIG. 3).

  The loan repayment history data 32 is data indicating the repayment history of past loans (for example, contract collection, overdue collection, partial advance collection, full advance collection, proxy payment collection).

  The market data 33 is data indicating market information (for example, market interest rate, interest rate volatility).

  The loan application data 34 is data indicating information applied by a loan applicant (that is, a debtor) at the time of applying for a loan. The loan application data 34 includes obligor attribute information and loan conditions.

  The past loan data 35 includes information on loans that have been loaned in the past (for example, loan number, debtor number, loan product name, execution date, loan date, guarantee company, outstanding balance, funding rate, interest rate spread, DTI (Debt to Income), LTV (Loan to Value), etc.) (see FIG. 4).

  The operation of the examination model creation device 1 of this embodiment will be described. FIG. 5 is a flowchart showing the procedure of the examination model creation process of the present embodiment.

  <S1-10> Using the obligor attribute data 31 and the loan repayment history data 32, the default model creation unit 10 creates a default model according to the obligor's attributes and loan conditions. The default model is to calculate the default rate using parameters (hereinafter referred to as “default model parameters”) corresponding to variables that affect the probability that the obligor will default in the future (hereinafter referred to as “default rate”). This is the calculation formula. For example, the default model is a Cox proportional hazard model obtained by a survival time analysis method.

Equation 1 shows an example of the default model. The default model parameters in the default model of Formula 1, the proportion of borrowing for the coefficient β _i (for example, the collateral valuation for the variables _{x i} representing the repayment ability of the debtor (LTV: Loan to Value) factor for the _β ltv, to the annual income the proportion of annual repayments: including coefficient β _dti) against (DTI Debt to Income), the baseline hazard function _h 0 of the number of elapsed months t from the loan execution (t), the a. That is, as shown in Equation 1, the default rate changes according to the elapsed time from loan execution and the numerical value representing the repayment capability based on the attribute.

  <S1-12> Using the loan repayment history data 32, the prepayment model creation unit 12 uses the loan repayment history data 32 for each repayment category (for example, full advance repayment, partial advance repayment) and products (fixed for all periods, fixed special contract term, and all Create a prepayment model for each period variation. A prepayment model is a prepayment rate that uses parameters (hereinafter referred to as “prepayment parameters”) that affect the probability that a debtor will execute prepayment in the future (hereinafter referred to as “prepayment rate”). Is a calculation formula for calculating.

Equation 2 shows an example of a prepayment model. The prepayment parameters in the prepayment model of Equation 2 are the probability that the debtor will execute full prepayment in the future (hereinafter referred to as “full prepayment rate”) Pa and the debtor will execute partial prepayment in the future. Pb (hereinafter referred to as “partially advanced repayment rate”) Pb. The full advance repayment rate Pa and partial advance repayment rate Pb are the first function f _i (t) with the elapsed months t as a variable and the interest rate difference d between the applicable interest rate and the market interest rate as a variable. 2 functions g _i (d) and a third function h _i (c) having the repayment calendar month c as a variable. That is, as shown in Equation 2, the prepayment rate changes according to the elapsed time from the loan execution, the applicable interest rate, the market interest rate, and the repayment calendar month.

In Equation 2, in many cases, the value of the first function f _i (t) increases as t increases (that is, the number of elapsed months increases), so the prepayment rate also increases. Also, the greater the interest rate difference d (for example, the lower the market interest rate), the greater the value of the second function g _i (d), so the prepayment rate also increases. In addition, since the value of the third function h _i (c) is high in the month where the number of relocations is large (generally in March), the prepayment rate in the month where the number of relocations is large also increases.

  <S1-14> The yield curve creation unit 14 creates a yield curve based on the market information (particularly, market interest rate) indicated by the market data 33.

  For example, the yield curve creation unit 14 calculates a discount factor for each grid point of the market interest rate, and creates a yield curve based on the calculated discount factor.

  <S1-16> The interest rate model parameter estimation unit 16 estimates the interest rate model parameter using the market data 33 (particularly, interest rate volatility).

  <S1-18> The fund procurement cost calculation unit 18 uses the loan application data 34, the payment model created in S1-12, the yield curve created in S1-14, and the interest rate estimated in S1-16. The funding cost is calculated by executing a simulation (for example, a Monte Carlo simulation) using the model parameters. That is, the fund procurement cost calculation unit 18 takes into account at least one of changes in prepayment over time and changes in market interest rates, and reflects the cost for the risk due to prepayment (hereinafter referred to as “prepayment risk”). A procurement cost (that is, a funding cost corresponding to a change in prepayment) is calculated. In particular, since the prepayment model created in S1-12 uses the elapsed time from loan execution and the market interest rate as variables, it is possible to consider changes in the prepayment according to the elapsed time and the market interest rate. become.

  Here, calculation of the funding cost according to the present embodiment will be described. FIG. 6 is a flowchart illustrating a procedure for calculating a funding cost (S1-18 in FIG. 5) according to the present embodiment.

  <S1-180> The funding cost calculation unit 18 creates an interest rate scenario.

  For example, the funding cost calculation unit 18 discretizes the hull white model and creates an interest rate scenario in which the short-term interest rate is a monthly interest rate. In this interest rate scenario, the discrete time is the first time at the end of the month following the evaluation reference date, and the unit time is monthly.

  <S1-182> The funding cost calculator 18 calculates a discount factor for each interest rate scenario. The discount factor is a coefficient indicating a discount rate for discounting future value into present value.

  <S1-184> The funding cost calculator 18 calculates a forward rate corresponding to the prepayment model. The forward rate for each interest rate scenario in the simulation indicates an index interest rate at a future time point.

  <S1-186> The fund procurement cost calculation unit 18 calculates the monthly interest payment rate at each repayment time for each interest rate scenario.

  For example, the funding cost calculation unit 18 adds the variable at the time of each repayment (elapsed months t, forward rate calculated in S1-184, assumed loan interest rate, and calendar month to the prepayment model created in S1-12. ) And calculate the monthly prepayment rate. The assumed loan interest rate is a value that is arbitrarily determined by the financial institution when creating the examination model.

  <S1-188> The funding cost calculation unit 18 executes cash flow development.

  For example, the funding cost calculation unit 18 executes the loan cash flow development for each interest rate scenario, using the monthly interest payment rate calculated in S 1-186. The funding cost calculator 18 carries out cash flow development for all periods of fixed-rate loans, but for loans for which the applicable interest rate until maturity has not yet been determined (for example, fixed-rate special housing loans). Executes the cash flow expansion until the interest rate is determined, and calculates the remaining principal cash flow at the final interest rate determination time.

  <S1-190> The funding cost calculator 18 calculates the funding cost.

  For example, the funding cost calculation unit 18 calculates the cash flow calculated in S1-188 (specifically, the principal repayment amount and the prepayment amount in each future repayment month) and the discount calculated in S1-182. The factor is used to calculate the market value of the loan (ie, the present value of the principal that is expected to be repaid in the future) for each interest rate scenario. The interest rate is calculated so that the average loan market value calculated for each interest rate scenario is equal to the loan principal. This calculated interest rate is the funding cost.

  <S1-20> The interest rate spread calculation unit 20 uses the default model created in S1-10, the prepayment model created in S1-12, the yield curve created in S1-14, and the S1-18 The calculated funding cost is used to calculate the interest rate spread as a function of default changes and prepayment changes. In particular, since the default model created in S1-10 uses the elapsed time from loan execution as a variable, it is possible to consider not only changes in prepayment but also changes in default over time.

  The calculation of the interest rate spread of this embodiment will be described. FIG. 7 is a flowchart showing a procedure for calculating the interest rate spread (S1-20 in FIG. 5) of the present embodiment.

  <S1-200> The interest spread calculation unit 20 determines the required rate of return (R (R is an arbitrary number)%).

  For example, the input device 5 receives an input of an instruction regarding the required rate of return. Next, the interest rate spread calculation unit 20 determines a required rate of return (R%) based on the instruction.

  <S1-202> The interest spread calculation unit 20 sets an internal variable. The internal variable is a variable used in the examination model creation apparatus 1 necessary for creating the examination model.

  For example, the interest rate spread calculation unit 20 acquires the loan application data 34 from the database 3 or the input device 5, and acquires the funding cost from the database 3 or the funding cost calculation unit 18. Next, the interest spread calculation unit 20 sets an internal variable based on the loan application data 34 and the funding cost.

  <S1-206> The interest rate spread calculation unit 20 calculates the forward rate. The forward rate is an index interest rate that starts on each loan repayment date indicated in the repayment schedule determined in S1-202.

  <S1-208> The interest spread calculation unit 20 sets an initial value (temporary interest spread) of interest spread (α (α is an arbitrary number)%).

  <S1-210> The interest spread calculation unit 20 sets an initial value “1” for the variable i.

  <S1-212> The interest spread calculation unit 20 determines whether or not the variable i is equal to or less than the number of lending months. When the variable i is equal to or less than the number of rented months (S1-212: Y), S1-214 is executed. When the variable i is larger than the number of rented months (S1-212: N), S1-226 is executed. The loan months are obtained from the execution date and the loan date.

  <S1-214> The interest spread calculation unit 20 calculates the contracted repayment amount on the i-th repayment date.

  For example, the interest rate spread calculation unit 20 determines the applicable interest rate by adding the interest rate spread (α%) set in S1-208 to the funding cost acquired in S1-202. Next, the interest spread calculation unit 20 calculates the principal repayment portion to be repaid on the i-th repayment date based on the applicable interest rate, the remaining principal amount, and the number of repayments. This principal repayment is the contracted repayment amount.

  <S1-216> The interest spread calculation unit 20 calculates the repayment amount before the deadline on the i-th repayment date.

  For example, the interest spread calculation unit 20 selects a prepayment model corresponding to the loan product from the prepayment models created in S1-12. Next, the interest spread calculation unit 20 sets the variable of the prepayment model as the number of months since execution, the applicable interest rate determined in S1-214, the forward rate calculated in S1-206, and the i-th repayment. Using the calendar month of the month to which the day belongs (hereinafter referred to as “current month”), the prepayment rate for the current month is calculated. Next, the interest spread calculation unit 20 calculates the remaining principal after repayment of the principal of the month (hereinafter referred to as “previous month”) to which the i−1th payment date belongs (information obtained in S1-220 described later), and this prepayment. Multiply by rate. The product of the remaining principal after repayment of the principal and the prepayment rate in the previous month is the prepayment amount before the deadline on the i-th repayment date.

  <S1-218> The interest spread calculation unit 20 calculates the default amount of the i-th repayment date.

  For example, the interest spread calculation unit 20 selects a default model corresponding to the loan condition from the default models created in S1-10. Next, the interest spread calculation unit 20 sets the variables (for example, DTI, LTV) representing the number of months elapsed since the execution and the debtor's repayment ability as the variables of the default model. Next, the interest spread calculation unit 20 calculates the default rate of the i-th repayment date based on the set default model variable. Next, the interest spread calculation unit 20 multiplies the remaining principal amount after repayment of the principal of the previous month by the default rate. The product of the remaining principal amount after the repayment of the previous month's principal and the default rate is the default amount on the i-th repayment date.

  <S1-220> The interest spread calculation unit 20 calculates the remaining principal after repayment on the i-th repayment date.

  For example, the interest rate spread calculation unit 20 calculates the principal repayment amount for the current month, the prepayment amount for the current month (information obtained in S1-216), and the default amount for the current month (from S1-216). And the information obtained in S1-218). The difference between the previous month's information (principal remaining after repayment of principal) and the current month's information (principal repayment amount, prepayment amount and default amount) is the remaining principal amount after repayment of principal in this month.

  <S1-222> The interest rate spread calculation unit 20 calculates the balance of the i-th repayment date using information on the balance of the loan. For example, the balance of the i-th repayment date is obtained from at least one of interest spread income, fee income, management cost, and default uncollected amount.

  For example, the interest spread calculation unit 20 multiplies the remaining principal after repayment of the principal of the previous month and the interest spread α. The product of the remaining principal after the repayment of the previous month and interest rate spread (α%) is interest rate spread income for the current month.

  Further, the interest spread calculation unit 20 multiplies the remaining principal after repayment of the principal in the previous month and the fee rate. The product of the remaining principal after the repayment of the previous month's principal and the fee rate is the fee income for the current month.

  Moreover, the interest spread calculation unit 20 multiplies the remaining principal after repayment of the principal of the previous month and the management cost rate. The remaining principal after the repayment of the principal in the previous month and the management cost ratio are the management costs for the current month.

  In addition, the interest spread calculation unit 20 multiplies the default amount of the current month by the assumed uncollected rate. The product of this month's default amount and the assumed uncollected rate is the default uncollected amount for the current month.

  <S1-224> The interest spread calculation unit 20 adds “1” to the variable i. When S1-224 ends, S1-212 is executed. That is, S1-214 to S1-222 are repeatedly executed until the variable i becomes equal to the number of rented months. And if the variable i becomes larger than the number of lending months, S1-226 is executed.

  <S1-226> The interest rate spread calculation unit 20 calculates the lifetime rate of return (V%) using the balance of the i-th repayment date calculated in S1-222. The lifetime rate of return is the average rate of return for the period from the time the loan is executed until the repayment is completed.

  For example, the interest rate spread calculation unit 20 subtracts the management cost for each month and the default uncollected amount for each month from the interest rate spread revenue for each month and the fee income for each month. The difference between the interest rate spread revenue and fee revenue and the management cost and default uncollected amount is the lifetime revenue amount.

Next, the interest spread calculation unit 20 divides the lifetime profit amount by the loan amount, and further divides by the loan years. That is, the lifetime rate of return is expressed by Equation 3.

  <S1-228> The interest rate spread calculation unit 20 determines whether the lifetime rate of return (V%) and the required rate of return (R%) substantially match. The interest rate spread calculation unit 20 determines that the lifetime rate of return (V%) and the required rate of return (R%) completely match, or the lifetime rate of return (V%) corresponds to the required rate of return (R%). When included in the predetermined allowable range, it is determined that the lifetime rate of return (V%) and the required rate of return (R%) substantially match. If the lifetime rate of return (V%) matches the required rate of return (R%) (S1-228: Y), the calculation of the interest rate spread ends. If the lifetime rate of return (V%) and the required rate of return (R%) do not match (S1-228: N), S1-230 is executed.

  <S1-230> The interest rate spread calculation unit 20 adjusts the interest rate spread (α%). When S1-230 ends, S1-210 is executed. That is, S1-210 to S1-230 are repeatedly executed until the lifetime rate of return (V%) matches the required rate of return (R%).

  Through S1-200 to S1-230, an interest rate spread (α%) is obtained such that the lifetime rate of return (V%) and the required rate of return (R%) substantially match.

  In the first example of S1-20, the interest rate spread calculation unit 20 uses the loan application data 34 and the required rate of return data in addition to the default model, the prepayment model, the yield curve, and the funding cost. Calculate interest rate spreads in response to default changes (eg, aging) and changes in prepayments.

  In the second example of S1-20, the interest rate spread calculation unit 20 uses the loan application data 34 and the required rate of return data in addition to the default model, the prepayment model, the yield curve, and the funding cost. Taking into account changes in defaults and changes in prepayment, the appropriate guarantee fee rate to be paid to the guarantee company is calculated.

  In the third example of S1-20, the interest rate spread calculation unit 20 uses the past loan data 35 and the obligor attribute data 31 in addition to the default model, the prepayment model, the yield curve, and the funding cost. S1-202 to 1-226 (however, instead of executing S1-208, use the interest rate spread in the database) to calculate the current lifetime rate of return for multiple past loans.

  According to the prior art, when creating a screening model, changes in defaults and changes in prepayment are not taken into consideration, and it has been difficult to set an appropriate interest rate for financial institution earnings. This is due to incorrect estimation of individual credit risk.

  In particular, since a complex algorithm is required to calculate the funding cost, the funding cost is calculated in consideration of changes in the prepayment according to the loan conditions of the applicant (especially fluctuations in market interest rates). This is a heavy burden for financial institutions. Therefore, in the prior art, it is difficult to calculate the funding cost in practice.

  Furthermore, according to the conventional technology, since it is impossible to accurately estimate the risk of bad debt, it is not possible to grasp the appropriate interest rate, so when setting a low interest rate to a customer with a high risk of bad debt, or the risk of bad debt Opportunity loss may occur due to the failure to present a reasonable interest rate (lower interest rate) to customers with low interest rates.

  In contrast, according to the present embodiment, the examination model is created taking into account changes in defaults and changes in prepayment, so that an appropriate interest rate for financial institution profits can be set without increasing the burden on financial institutions. It can be easily calculated. Thereby, the profit of a financial institution can be improved.

  At least a part of the examination model creation apparatus 1 according to the present embodiment may be configured by hardware or software. When configured by software, a program that realizes at least a part of the functions of the examination model creation apparatus 1 may be stored in a recording medium such as a flexible disk or a CD-ROM, and read and executed by a computer. The recording medium is not limited to a removable medium such as a magnetic disk or an optical disk, but may be a fixed recording medium such as a hard disk device or a memory.

  In addition, a program that realizes at least a part of functions of the examination model creation device 1 according to the present embodiment may be distributed via a communication line (including wireless communication) such as the Internet. Further, the program may be distributed in a state where the program is encrypted, modulated or compressed, and stored in a recording medium via a wired line such as the Internet or a wireless line.

  In addition, this invention is not limited to embodiment mentioned above, It deform | transforms and implements a component in the range which does not deviate from the summary. Various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the above-described embodiments. For example, you may delete a some component from all the components shown by embodiment mentioned above. Furthermore, constituent elements over different embodiments may be appropriately combined.

DESCRIPTION OF SYMBOLS 1 Review model creation apparatus 3 Database 5 Input apparatus 7 Output apparatus 9 Communication apparatus 10 Default model creation part 12 Prepayment model creation part 14 Yield curve creation part 16 Interest rate model parameter estimation part 18 Funding cost calculation part 20 Interest rate spread calculation part

Claims (11)

A default model creation section for creating a default model for loan applicants;
A prepayment model creation unit for creating a prepayment model of the applicant;
Using the prepayment model, a funding cost calculation unit that calculates a funding cost according to a change in the prepayment,
An interest rate spread calculation unit for calculating an interest rate spread according to a change in default and a change in the payment using the default model, the prepayment model, and the funding cost;
An examination model creation device characterized by comprising: The funding cost calculation unit
Calculate the prepayment rate according to the interest rate scenario,
The examination model creation apparatus according to claim 1, wherein the funding cost is calculated using the prepayment rate. The funding cost calculation unit
Using the prepayment rate, execute the cash flow development of the loan for each interest rate scenario,
Calculate the present value of the loan based on the execution result of the cash flow expansion,
The examination model creation device according to claim 2, wherein an interest rate at which a present value of the loan is equal to a loan principal is calculated as the funding cost. The interest rate spread calculator
Determine the required rate of return,
Calculate the lifetime rate of return, which is the average rate of return from loan execution to repayment,
The examination model creation device according to any one of claims 1 to 3, wherein an interest rate spread is calculated when the required rate of return and the lifetime rate of return substantially coincide. The interest rate spread calculator
Using the default model and the prepayment model, execute cash flow expansion,
Calculate the balance of each repayment date,
Calculate the lifetime rate of return based on the balance,
The examination model creation device according to claim 4 which computes an interest rate spread when said demanded return rate and said lifetime return rate substantially correspond. The interest rate spread calculator
Set a temporary interest spread,
Using the provisional interest rate spread, calculate the contracted repayment amount for each repayment date in the repayment schedule,
Calculate the prepayment amount for each repayment date,
Calculate the default amount for each repayment date,
Calculate the remaining principal after repayment for each repayment date,
Using the temporary interest rate spread, calculate the balance for each repayment date,
Based on the balance, calculate the lifetime rate of return,
6. The examination model according to claim 5, wherein if the required rate of return and the lifetime rate of return do not substantially match, the provisional interest rate spread is changed and the contracted repayment amount and the balance are recalculated. Creation device.   The said interest rate spread calculation part calculates the lifetime rate of return regarding a plurality of past loans using the default model, the prepayment model, and information about past loans. Screening model creation device.   The examination model creation device according to any one of claims 1 to 7, wherein the default model uses an elapsed time from loan execution as a variable.   The examination model creation device according to claim 1, wherein the prepayment model uses at least one of an elapsed time from loan execution and a market interest rate as a variable. On the computer,
Creating a default model for loan applicants;
Creating a prepayment model for the applicant;
Using the prepayment model to calculate funding costs in response to changes in the prepayment;
Using the default model, the prepayment model, and the funding cost to calculate an interest rate spread according to a change in default and a change in the prepayment;
A computer program that executes Computer
Creating a default model for loan applicants;
Creating a prepayment model for the applicant;
Using the prepayment model to calculate funding costs in response to changes in the prepayment;
Using the default model, the prepayment model, and the funding cost to calculate an interest rate spread according to a change in default and a change in the prepayment;
Execute the examination model creation method.

Images