JP2000020603A - Method and device for calculating risk and storage medium with risk calculation program stored therein - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a calculation method for quickly calculating value at risk and an earning at risk by calculating the value at risk and the earning at risk, based on the detailed information of a certain item and an item rate scenario. SOLUTION: A market DB preparation program realized a date (here-after referred to as an interest changing date) for reconsidering interest in the future based on the information of a details DB and applies interest, corresponding to an interest changing date in an item interest rate scenario DB. Then market data are prepared and stored in a market DB. A VAR/EAR calculation program calculates the current price or the like, before the change of an interest scenario from the information of the market DB and the information of the interest scenario and stores the calculated result in a simulation result DB. Since the value at risk and an earning at risk calculated by this Monte Carlo simulation are highly efficient, processing speed can be improved, while the accuracy of calculation is maintained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a financial institution that creates market data (interest rate scenario) that changes under variance / covariance obtained from historical data and uses this to generate portfolio value fluctuations. Value-at-risk (a risk indicator stochastically expressing the maximum loss that can occur in a portfolio with a certain probability during the holding period) and earning-at-risk (a period expected to be realized in the future) The present invention relates to a method and an apparatus for performing a risk calculation method for calculating an expected value of profit and loss and a probability distribution), and a storage medium storing a program for executing the method.

[0002]

2. Description of the Related Art Conventionally, financial institutions have been using JPorg.
Calculate the value-at-risk using the risk metrics method published by
The value at risk and the learning at risk are calculated by simulation. The method of calculating the value-at-risk by Monte Carlo simulation is described in Muta: "Financial Monte Carlo", issued by Hyundai Sales Co., May 14, 1996. The method of calculating the earning-at-risk is described by Okubo: "Earning-at-risk.・ Risk ”, Study Group on Financial Affairs, May 26, 1997
It is specifically disclosed in the date issued.

[0003]

The above-described calculation of value-at-risk by the risk metrics method has a problem that there is a risk (non-linear risk) that cannot be grasped. For value-at-risk and learning-at-risk calculated by Monte Carlo simulation, it is possible to grasp risks (nonlinear risks) that cannot be grasped by risk metrics and increase the number of assets to be covered. On the other hand, the amount of calculation increases exponentially in other calculation methods, but increases only proportionally, which is efficient. However, value-at-risk and learning
There is no description of how to calculate at risk. Also, Monte Carlo simulation
Another problem is that it takes a lot of time.

[0004] It is an object of the present invention to provide a calculation method and apparatus for calculating a value-at-risk or an learning-at-risk at a high speed by calculating by Monte Carlo simulation, and a storage storage medium embodying the present invention. It is in.

[0005]

In order to achieve the above object, the present invention provides a statement DB in which statement information of each item (regular deposit, reserve period, etc.) traded by a financial institution or the like is stored.
And a subject interest rate scenario DB that stores a scenario in which the user predicts what value the interest rate will be in the future for each subject in each month, and is used when calculating the present value. Interest rate scenario D in which the discount rate is stored
Suppose B has already been given. Based on the information in the detail DB by the market DB creation program, the year when the interest rate will be revised in the future (hereinafter referred to as the renewal date) is grasped, and the interest rate corresponding to the renewal date in the interest rate scenario DB is applied. I do. Thereafter, a market is created and stored in the market DB. VAR / EA
The R calculation program calculates the market price before the change in the interest rate scenario from the information in the market DB and the information on the interest rate scenario, and stores it in the simulation result DB.

The interest rate path creation program creates an interest rate scenario and an interest rate scenario for the number of simulations designated by the user and stores them in the work area.
Thereafter, the VAR / EAR calculation program creates a market after the interest rate scenario is changed based on the item interest rate scenario and the interest rate scenario stored in the work area and information stored in the market DB, and calculates a market value after the interest rate scenario is changed. .
Thereafter, a statistical calculation is performed using the market price before the interest rate scenario change or the market price after the interest rate scenario change, and the result is stored in the simulation result DB.

[0007] As a result, the value-at-risk and the learning-at-risk calculated by the Monte Carlo simulation are obtained by grasping risks (non-linear risks) that cannot be grasped by risk metrics, and increasing the number of assets to be covered. On the other hand, the amount of computation increases exponentially with other methods, but only proportionally,
Efficient.

Further, the present invention provides a method for performing the above method.
Also provided is a method for implementing a Monte Carlo simulation for improving the processing speed while maintaining the calculation accuracy.

[0009] Based on the information in the detail DB in the market DB program, the date when the interest rate will be revised in the future (interest renewal date) is grasped,
The interest rate corresponding to the renewal date of the interest rate scenario DB is applied. After that, a market is created, and those having the same renewal date are aggregated and stored in the market DB (market DB by profit and renewal month).
The VAR / EAR calculation program calculates the market price before the change in the interest rate scenario from the information in the market DB and the information on the interest rate scenario, and stores it in the simulation result DB.

The interest rate path creation program creates an interest rate scenario and an interest rate scenario for the number of simulations specified by the user and stores them in the work area.
After that, based on the interest rate scenario and interest rate scenario stored in the work area by the VAR / EAR calculation program and the information stored in the market DB (market DB by month of renewal), market after the change of the interest scenario (market by month of renewal) And calculate the market value after the interest rate scenario changes. afterwards,
Statistical calculation is performed using the market price before the interest rate scenario change or the market price after the interest rate scenario change, and the result is stored in the simulation result DB.

As a result, it is possible to realize a Monte Carlo simulation for improving the processing speed while maintaining the calculation accuracy.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a method for implementing a Monte Carlo simulation according to a first embodiment of the present invention will be described with reference to FIG.

The Monte Carlo simulation to which the present invention is applied includes a display 100, a keyboard 101,
Central processing unit (CPU) 102, CD-ROM driver 120, item DB 103, interest rate scenario DB 10
4. Storage device 10 for storing market DB 105, interest scenario DB 106, and simulation result DB 107
8. It is composed of the main memory 110. The CPU 109, the CD-ROM driver 120, the storage device 105, and the main memory 110 are connected to the bus 109.

The main memory 110 stores a control program 11
1, a market DB creation program 112, an interest rate pass creation program 112, a VAR / EAR calculation program 114, and a work area 115 are stored. The commercial DB creation program 112, the interest rate pass creation program 113, and the VAR / EAR calculation program 114 are stored in a storage medium such as a portable CD-ROM or a floppy disk, and are stored in the storage device 108 at the time of installation. When the apparatus is started, the market DB creation program 112, the interest rate pass creation program 113, and the VAR /
The EAR calculation program 114 is read from the storage device 108 by the control program 111 and stored in the memory 110.

Market DB creation program, interest rate pass creation program 113, and VAR / EAR calculation program 1
Step 14 is executed by the CPU 102 in response to an instruction from the keyboard 101 by the user.

FIG. 3 shows an example of the detailed configuration of the information in the detail DB 103 stored in the storage device 108. Each of the items (ordinary accounts,
It is assumed that the information is provided as a detail DB based on the detail information of the funding period.

The management store name 301 which is information on the sales office where the transaction was made, and the customer name 30 which is information on the customer who made the transaction
2. Transaction number 303 assigned to each business unit,
Account code 304, which is information for aggregating the details into certain items (ordinary deposit, reserve period, etc.), reference date 305, which is information on when the details are, and deposits with which the customer has made a transaction Day 306, about a fixed date 307, which is information on the maturity date of the transaction, a contracted interest rate 308, which is information on the interest rate given to the customer by the financial institution, between the head office and the branch attached inside the financial institution (the sales office and the head office). Partition rate 30 which is the interest rate
9. A contracted interest rate review cycle 310 which is information on a cycle (period) for reviewing the interest rate of a variable subject (regular fluctuation, variable loan, etc.), and a current balance 31 which is a current balance of each item.
1. Fixed / floating category 312, which is information on whether the transaction statement is fixed or variable, interest payment cycle (period)
Interest payment cycle 313 is stored.

The specification DB 103 is input data for creating the market DB 105 by the market DB creation program 112.

FIG. 4 shows an example of a detailed configuration of information of the subject interest rate scenario DB 104 stored in the storage device 108. The future interest rate for each subject is shown on a monthly basis. It is assumed that the predicted scenario has already been created and given by the user.

Subject code 4 which is information corresponding to the subject name
01. Is the interest rate scenario that corresponds to the new transaction the market (the assets and liabilities currently held will decrease in the future. )), A new market segment 402 which is information on whether the subject interest rate scenario corresponds to, a year 403 which is information on the subject interest rate scenario for which year, a future reference rate for each subject (transactions with customers). Rate that
The reference rate 40 which is information on the normal contract rate)
4, 406, 408, 410, 412, 414, 41
6, 418, 420, 422, 424, 426, and partition rates 405, 407, 409, 411 which are information on future partition rates for each subject (rates for partitioning the inside of the bank separately from the actual customer rates). , 413, 415,
417, 419, 421, 423, 425, and 427 are stored.

The subject interest rate scenario DB 104 is input data for creating the market DB 105 by the market DB creation program 112.

FIG. 5 shows an example of the detailed configuration of the information of the market DB 105 stored in the storage device 108. The detailed DB 103 and the subject interest rate scenario 104 are input and are created by the market DB creation program 112. .

Subject code 5 which is information corresponding to the subject name
01, fixed / variable classification 502, which is information on fixed or variable subjects, and cash flows generated in any fiscal year (Sumitomo Trust & Banking Market Finance Department, Derivative Keywords 25
0, Study Group on Financial Affairs, P103, July 1, 1995
(Refer to the 8th day), the year 503 which is information as to whether it is information, and the base date 504 which is information as to when the detailed information at the time is aggregated for each month, and when the interest rate is revised (interest rate review) in the future. 505, the principal cash flow 50, which is the principal cash flow information.
6,509,512,515,518,521,52
4, 527, 530, 533, 536, and 539; cash flow information 507, 510, 513, 516, 519, and 5, which are cash flow information of interest.
22, 525, 528, 531, 534, 537, 54
0, partition interest cash flows 508, 511, 514, 517, which are cash flow information of the partition interest;
520, 523, 526, 529, 532, 535, 5
38 and 541 are stored.

The commercial DB 105 is input data for creating a simulation result DB by the VAR / EAR calculation program 113.

FIG. 6 shows an example of the detailed configuration of the information of the interest rate scenario DB 106 stored in the storage device 108. The discount rate used when calculating the present value has already been created and given by the user. It is assumed that

The year 60, which is information on what year it is
1. Discount factor for each month (by Sumitomo Trust and Banking, Market Finance Department, Derivatives Keyword 250, Study Group on Financial and Fiscal Conditions, pp. 244-245, July 18, 1995)
DF rate 602, 60 which is the rate information
3, 604, 605, 606, 607, 608, 60
9, 610, 611, 612, and 613 are stored.

This interest rate scenario DB 106 stores VAR /
This is input data for creating the simulation result DB 107 by the EAR calculation program 114.

FIG. 7 shows an example of a detailed configuration of the information of the simulation result DB 107 stored in the storage device 108.
B106, the interest rate scenario created by the interest rate creation program 112 and the subject interest rate scenario
Created by the AR / EAR calculation program 113.

Subject code 7 which is information corresponding to the subject name
01, a customer / partition classification 70 which is information on the customer / partition
2. Market value before interest rate fluctuation, which is the value of the market price before interest rate fluctuations (re-evaluating cash flows based on the interest rate situation) 7
03, the expected value 704 of the market price after the market price fluctuation, which is the average value of the market price after the interest rate fluctuates, the market price change standard deviation 705, the standard deviation of the market price after the interest rate fluctuates, the left tail confidence interval 3σ (Σ is the standard deviation) the market price change width (3σ left tail) 706 which is the value of the market price, the market price change width (2σ left tail) 707 which is the market price value of the left foot confidence interval 2σ, and the left foot confidence interval 1.6
Market price change width which is the market price value of 5σ (1.65σ left tail) 7
08, the market price change width (3σ right tail) 709 which is the value of the market price of the right tail confidence interval 3σ, the market price change width (2σ right tail) 710 which is the market price value of the right foot confidence interval 2σ, and the right tail confidence interval 1. 6
Market price change width which is the market price value of 5σ (1.65σ left tail) 7
11 is stored.

This simulation result DB 107 is
The result data calculated by the VAR / EAR calculation program 114 is stored.

In such a configuration, the Monte Carlo
The simulation realizing process will be described with reference to FIGS.

FIG. 2 is a flowchart showing an example of a risk management program to which the risk calculation method of the present invention is applied.

First, an outline of the flow will be described. First, in a statement extracting step 100, a statement corresponding to the same subject is extracted from the statement DB 103 stored in the storage device.

In the subject interest rate scenario (before change) market creation step 105, the extracted DB detailed information and the subject interest rate scenario DB 104 are input and the market creation DB 11
In step 2, a market for the interest rate scenario (before change) is created and stored in the market DB 105.

In the interest rate scenario (before change) market value calculation step 110, the interest rate scenario is calculated by the VAR / EAR calculation program 114 using the market DB 105 and the interest scenario DB 106 created in the subject interest rate scenario (before change) market creation step 105 as inputs. (Before change) The present value is calculated and stored in the simulation result DB 107.

In the interest rate pass generation step 115, the control program 111 sets the interest rate scenario (after the change) or the subject interest rate scenario (after the change) for the number of simulations input from the keyboard 101 into the interest rate pass creation program 1
13 and stored in the work area 115.

In the fixed / variable determination step 120, the VAR / EAR calculation program 114 determines whether the subject is a fixed subject or a variable subject from the market DB 105 of the created specification.

In the case of a fixed subject, in the interest rate scenario (after change) market value calculation step 125, the market scenario DB 105 and the interest rate scenario (after change) created in the interest rate pass generation step 115 are input and the VAR / EAR calculation program 114 is used. The interest price scenario (after the change) is calculated and stored in the work area 115.

In the case of a variable subject, in the subject interest rate scenario (after change) market creation step 130, the VAR / EAR calculation program 114 receives the subject interest rate scenario (after change) created by the market DB 105 and the interest rate pass generation step 115 as input. Calculates the execution of the subject interest rate scenario (after change), stores it in the work area 115, and performs the interest rate scenario (after change) market value calculation step 125.

Simulation number determination step 135
Now, it is determined whether the control program 111 has repeated the fixed / change determination step 120, the interest rate scenario (after change) market value calculation step 125, and the interest rate scenario (after change) market creation step 130 for the number of simulations input from the keyboard 101. The determination is performed by the VAR / EAR calculation program 114.

In the detailed result calculation step 140, the expected value or the standard value is calculated using the interest price scenario (before change) and the interest price scenario (after change) obtained by the control program 111 being repeated for the number of simulations inputted from the keyboard 101. VAR / EAR calculation program 114
Is calculated and stored in the work area.

In the number-of-details determination division step 145, the VAR / EAR calculation is performed to determine whether the simulation result calculation step 140 has been repeated from the subject interest rate scenario (before change) market creation step 105 for the number of details extracted in the detail extraction step 100. The determination is made by the program 114.

Finally, in the simulation result calculation step 150, the VAR / EAR is calculated based on the expected value, the standard deviation, etc. calculated for each specification in the specification result calculation step 140.
The calculation program 114 calculates expected values, standard deviations, and the like for each subject, and stores them in the simulation result DB 107.

First, in the statement extracting step 100, the same item (ordinary deposit,
The details corresponding to the funding period are extracted based on the account code 304 based on the correspondence between the account and the account code defined by the user in advance.

Thereafter, in the subject interest rate scenario (before fluctuation) market creation step 105, the base date 305, deposit date 306, regular regular date 307, contracted interest rate review cycle 310, Balance 31
1. Interim payment cycle 313 and reference rates 404, 406, 408, 41 of the interest rate scenario DB 104
0, 412, 414, 416, 418, 420, 42
2, 424, 426 and partition rates 405, 407, 40
9, 411, 413, 415, 417, 419, 42
With the input of 1, 423, 425, and 427 as inputs, the market DB creation program 112 calculates principal cash flows, customer cash flows, and partition cash flows of the subject interest rate scenario (before change). Then, each cash flow of the calculated item interest rate scenario (before change) is stored in the market DB 105. The account code corresponding to the account item code is described in the account code 501 of the market DB 105, the fixed / variable category 312 of the detail DB 103 is defined in the fixed / variable category 502, the year in which the cash flow occurred in the year 503, and the base year. The date DB 504 contains the statement DB1
03 as the base date 305, and the principal cash flows 506, 509, 512, 515, 51 of each month.
8, 521, 524, 527, 530, 533, 53
6, 539, cash flow information on interest 507, 510, 513, which is cash flow information on interest on interest;
516, 519, 522, 525, 528, 531, 5
34, 537, 540, partition interest cash flows 508, 51 which are cash flow information of partition interest.
1, 514, 517, 520, 523, 526, 52
9, 532, 535, 538, and 541 store the calculated cash flows in the market DB 105.

Next, in the interest rate scenario (before change) market value calculation step 110, the subject interest rate scenario (before change) market creation step 105 is created and stored in the market DB1.
The VAR / EAR calculation program 11 receives the principal, reference and partition cash flows of 05, and the discount factor as information of the interest rate scenario DB 106 as inputs.
In step 4, the present value of the interest rate scenario (before change) is calculated.

Next, in the interest rate pass generation step 115,
Control program 1 from the interest rate pass creation program 113
11 is a Vasisek model (Masaaki Kijima, “Introduction to Finance Engineering Part 2”, Nikkagiren P11
4 to P121, September 14, 1995), and creates an interest rate scenario (after fluctuation) or a subject interest rate scenario (after fluctuation) and stores it in the work area 115.

Next, in the fixed / variable determination step 120, whether the extracted detail is fixed or variable is determined by V
Market DB 105 by AR / EAR calculation program 114
Is determined based on the fixed / variable classification 502 of.

If the extracted subject is a fixed subject, the following processing is performed.

In the case of fixed subjects, the contracted interest rate 308,
Since the partitioning rate 309 is fixed and it is not necessary to review the interest rate, in the interest rate scenario (after fluctuation) market value calculation step 125, the yearly 503 of the market DB 105 and the principal cash flows 506, 509, 512 of each month are set. , 51
5, 518, 521, 524, 527, 530, 53
3, 536, 539, interest cash flows 507, 510, which are cash flow information of interest.
513, 516, 519, 522, 525, 528, 5
31, 534, 537, 540, the partition interest cash flow 50 which is the cash flow information of the partition interest
8, 511, 514, 517, 520, 523, 52
6, 529, 532, 535, 538, and 541 and the interest rate scenario (after change) created in the interest rate pass generation step 115 are input to calculate the current market value of the interest rate scenario (after change).

Next, a process in the case where the extracted subject is a variable subject will be described.

In the case of a variable subject, in the subject interest rate scenario (after fluctuation) market creation step 130, the contracted interest rate 30
8. The applicable rate is fixed at the partition rate 309, and there is no need to review the interest rate (part where the interest date has not arrived),
The VAR / EAR calculation program 114 creates the cash flow of the item interest rate scenario (after fluctuation) separately from the portion where the applied interest rate is not fixed (the portion where the interest rate is due) when the interest rate is changed in the future.

First, the principal cash flow, the customer cash flow, and the partition cash flow in the portion where the applied interest rate is fixed and the interest rate does not need to be reviewed (the portion where the interest date has not yet arrived) are fluctuated. There is no. Principal cash flow, reference cash flow, and partition cash flow of the portion where the interest rate does not need to be reviewed (the portion where the interest date has not yet arrived) are calculated based on the deposit date 306, the contracted interest rate review cycle 310, and the interim interest payment in the detail DB 103. Cycle 313
I understand more.

However, the portion where the applicable interest rate has not been determined after the future interest rate renewal (interest date portion), that is, the target cash flow and the partitioning cash flow other than the cash flows for which the applied interest rate has been determined are Reference rate of the interest rate scenario for the month (prior to change) for future renewal of the month 40
4, 406, 408, 410, 412, 414, 41
6, 418, 420, 422, 424, 426 and partition rates 405, 407, 409, 411, 413, 4
15, 417, 419, 421, 423, 425, 42
7 and the reference rate 404 of the interest rate scenario (after change),
406, 408, 410, 412, 414, 416, 4
18, 420, 422, 424, 426 and partition rates 405, 407, 409, 411, 413, 415,
417, 419, 421, 423, 425, and 427 are obtained, and the cash flow of the subject interest rate scenario (after fluctuation) is calculated based on the reference and partition cash flows stored in the market DB 105 and the ratio calculated above. Create The future renewal month is 3
06, the contracted interest rate review cycle 310, and the interim interest payment cycle 313.

Thereafter, in an interest rate scenario (after fluctuation) market value calculation step 125, the cash flow created in the subject interest rate scenario (after fluctuation) market creation step 130 and the interest rate scenario ( The VAR / EAR calculation program 114 calculates the current market value of the interest rate scenario (after the change) by using the (after change) as an input.

Next, in a simulation number determination step 135, the control program 111
The VAR / EAR calculation program 114 determines whether or not the fixed / change determination step 120, the interest rate scenario (after change) market value calculation step 125, and the interest rate scenario (after change) market creation step 130 have been repeated for the number of simulations input from judge.

Next, in a detailed result calculation step 140,
The interest price scenario (after change) is calculated using the interest price scenario (after change) and the interest price scenario (before change) calculated in the interest price scenario (before change) market value calculation step 110, which are repeatedly calculated for the number of simulations. The expected value, standard deviation of market price change, and range of market price change (3 standard deviations, 2 standard deviations, 1.65 standard deviations) are calculated.

Next, in the number-of-details determination step 145, the number of details extracted in the detail extraction step 100 is calculated.
The VAR / EAR calculation program 114 determines whether the simulation result calculation step 140 is repeated from the subject interest rate scenario (before change) market creation step 105.

Finally, in the simulation result calculation step 150, the item code 701 of the simulation result DB 107 is the item code extracted in the detail extraction step 100, and the customer / partition category 702 is for the customer cash flow. Information on whether it is a count or a count for partition cash flows is based on the market value before interest rate fluctuation 703.
Is the total of the present value of the interest rate scenario (before change) calculated for each item calculated in the interest price scenario (before change) step 110 and the expected value 704 after the interest rate change is calculated in the detail result calculation step 140 The total of the interest price scenarios (after fluctuations) for each item and the standard value change standard deviation 705 are the total of the standard deviations for each item calculated in the item result calculation step 140, and the market value change ranges 706, 707, 70
8, 709, 710, and 711 show the total of the market value change widths for each item calculated in the item result calculation step 140 in VA.
It is calculated by the R / EAR calculation program 114 and stored.

The information in the simulation result DB 107 is used as an index for grasping the risk of each financial institution and how to cope with the risk.

Hereinafter, a specific example will be described.

The detail extraction step 100 allows the detail DB1
03, the fixed / variable division 312 of the item is fixed, the base date 305 = 1997/7/31, the deposit date 3
06 = 1997/5/31, about regular date 307 = 1999
/ 5/31, contracted interest rate 308 = 0.5%, partition rate 3
09 = 0.8%, current balance 311 = 1,000,000
Yen, interim payment cycle 313 = 12 months.

In the interest rate scenario (before change) market creation step 105, the interim interest payment cycle 313 is 12 months, the deposit date 306 is May 31, 1997, and the regular
7 from May 19, 1999 to May 31, 1998
The value of the reference and partition interest cash flows occurring on May 31, 1999 is determined.

The value of the reference and partition cash flows generated on May 31, 1998 is calculated as follows:
Since the interest cash flow is up to 1, the interest cash flow for the customer = the current balance 311 x the contracted interest rate 308 x (the number of days from the base date 305 to the date of interest generation) / 365 = 1,000,000 x 0.5 / 100 x (days from July 31, 1997 to May 31, 1998) / 3 365 partition interest cash flow = current balance 311 x partition rate 309 x (days from base date 305 to date of interest occurrence) /365=1,000,000×0.8/100×(days from 1997/7/31 to 1998/5/31) / 365.

The value of the reference and partition cash flows generated on May 31, 1999 is based on the previous interim interest payment date of 19
Next interim payment date from May 31, 1998 May 3, 1999
Since the interest cash flow is up to 1, interest cash flow to the customer = current balance 311 x contracted interest rate 308 x (days from the previous interim interest payment date to the next interim interest payment date) / 365 = 1,000,000 x 0. 5/100 x (days from May 31, 1998 to May 31, 1999) / 3 65 Partition interest cash flow = current balance 311 x partition rate 308 x (from the previous interim payment date to the next interim payment date) Days) /365=1,000,000×0.8/100× (days from 1998/5/31 to 1999/5/31) / 365. In addition, about 307 on 1999/5
/ 31 has a principal cash flow of 1,000,000
0 also occurs.

The cash flow information obtained as described above is stored in the market DB 105.

Next, the details are extracted from the details DB 103 by the details extracting step 100, and the fixed / variable category 312 of the details is extracted.
Has already changed, the contracted interest rate 308 and the partition rate 30
In 9 there is a portion where the applied interest rate is fixed and there is no need to review the interest rate (part where the interest rate date has not yet arrived), and there is a portion where the interest rate has not been fixed due to future interest rate renewal (part where the interest rate date has arrived).

First, in the subject interest rate scenario (before change) market creation step 105, the principal cash flow of the portion where the applied interest rate is fixed and the interest rate does not need to be reviewed (the portion where the interest date has not arrived) is calculated. Client cash flow, partition cash flow, the base date 305,
Calculated from the deposit date 306, the regular period 307, the contracted interest rate 308, the partition rate 309, the contracted interest rate review cycle 310, the current balance 311, and the interim interest payment cycle 313.

For example, the reference date 305 = 1997/7
/ 31, deposit date 306 = 1997/5/31, about regular date 307 = 1999/5/31, contracted interest rate 308 = 0.5
%, Partition rate 309 = 0.8%, contracted interest rate review cycle 310 = 12 months, current balance 311 = 1,000,
000 yen, interim payment cycle 313 = 12 months. The contract rate review cycle 310 is 12 months, and the interim payment cycle 313 is 12 months,
The reference and interest cash flows generated in (1) are fixed and calculated as follows.

The value of the reference and partition cash flows generated on May 31, 1998 is calculated as follows.
Interim payment date from 997/7/31 1998/5/31
Since the interest cash flow until then, the interest cash flow for the customer = current balance 311 x contracted interest rate 308 x (days from the base date 305 to the date of interest generation) / 365 = 1,000,000 x 0.5 / 100 x (days from July 7, 1997 to May 31, 1998) / 3 65 Partition interest cash flow = current balance 311 x partition rate 309 x (days from base date 305 to date of interest) / 365 = 1,000,000 x 0.8 / 100 x (days from 1997/7/31 to 1998/5/31) / 3365.

Next, in step 105 for creating a subject interest rate scenario (before change), the principal cash of the portion where the interest rate will be changed in the future and the applied interest rate is not fixed (interest date portion) will be described.
The cash flow, customer cash flow and partition cash flow are calculated using the base date 305, deposit date 316,
About the regular date 307, the contracted interest rate review cycle 310, the current balance 311, the interim interest payment cycle 313, and the reference rate 40 of the subject interest rate scenario DB 104 in the storage device 32
4, 406, 408, 410, 412, 414, 41
6, 418, 420, 422, 424, 426 and partition rates 405, 407, 409, 411, 413, 41
5, 417, 419, 421, 423, 425, 427
Calculate from

For example, in the same manner as described above, in the case of a variable period subject, the base date 306 = 1997/7/31, the deposit date 306 = 1997/5/31, and the approximate period 307 = 19
99/5/31, contracted interest rate review cycle 310 = 12
Month, current balance 311 = 1,000,000 yen, interim payment cycle = 12 months. The reference and cash interest cash flows generated on May 31, 1999 are calculated from the reference rate 406 and the cash flow rate 407 of the 1998/5 subject interest rate scenario DB 104 in which the contracted interest rate review cycle 310 is 12 months or more in the future. Is done.
1998/5 reference rate = 0.6, partition rate = 0.
9 is assumed. The value of the reference and partition cash flows generated on May 31, 1999 is based on the previous interim interest payment date of 1998.
Interest cash flow from / 5/31 to the next interim payment date 1999/5/31, so interest cash flow = current balance 311 x reference rate x (from last interim payment date to next interim payment date) Days) /365=1,000,000×0.6/100× (days from May 5, 1998 to May 31, 1999) / 3 65 Partition interest cash flow = Current balance 311 × Partition rate × ( The number of days from the previous interim payment date to the next interim payment date) / 365 = 1,000,000 x 0.9 / 100 x (the number of days from 1998/5/31 to 1999/5/31) / 365; Is calculated. In addition, about 307 on 1999/5
/ 31 has a principal cash flow of 1,000,000
0 also occurs.

After that, the calculated cash flows are stored in the market DB 105.

In the year 503 of the market DB 105, the year 1998 is changed to 2,0 by the target cash flow 510 in May of that year.
00 is 4,00 in May cash flow 511
Suppose that 0 is stored, 3,000 is stored in the reference cash flow 510 in May, 1999, and 5,000 is stored in the partition cash flow 511 in May in the year 503 in the year.

In the interest rate scenario DB 106, the year 60
1 to 1998 is 99.5 for the May discount factor 603 for the year, 1999 to 991 for the May discount factor 603 for the year 601.
It is assumed that 0 is stored.

Then, in an interest rate scenario (before change) market value calculation step 110, the present value of the interest rate scenario (before change) is calculated as follows.

Present value of interest rate scenario (before change) (reference) = Cash flow 510 for fiscal year 503 in May and reference to cash flow 510 × Discount factor for fiscal year 503 in May and 603 + fiscal year 503 for 1999 May reference cash flow 510 x year 503 is 1999 and May discount factor 603 = 2,000 x 99.5 / 100 + 3,000 x 99.0 / 100 Present value of interest rate scenario (before change) (partition) ) = Cash flow 511 for fiscal 503 in May and May 511 × Discount factor for fiscal 503 in May and May 603 + Cash flow for fiscal 503 in May and May 511 × Fiscal 503 for May and 1999 Discount factor 603 = 4,000 x 99.5 / 10 + 5,000 × 99.0 / 100 Thereafter, in the interest rate path generation step 115, Vasic
The interest rate scenario (after fluctuation) or the subject interest rate scenario (after fluctuation) is calculated by a method such as the ek model.

If the fixed / floating category 502 of the market DB 105 is fixed at the fixed / fluctuating determination step 120, the contracted interest rate 308 and the partition rate 309 have already been determined, and there is no need to review the interest rate. The present value after the change in the interest rate scenario is calculated by multiplying the cash flow data stored in the market DB 105 by the discount factor value of the corresponding interest rate scenario (after the change).

In the year 503 of the market DB 105, 1998 is 2,0 in the target cash flow 510 in May of that year.
00 is 4,00 in May cash flow 511
Suppose that 0 is stored, 3,000 is stored in the reference cash flow 510 in May, 1999, and 5,000 is stored in the partition cash flow 511 in May in the year 503 in the year.

In the interest rate pass generation step 115,
It is assumed that the discount factor in May 1998 of the interest rate scenario (after fluctuation) created by a method such as the Vasisek model is 99.2, and the discount factor in May 1999 is 98.8.

Then, in an interest rate scenario (after change) market value calculation step 125, the present value of the interest rate scenario (after change) is calculated as follows.

Present value of interest rate scenario (before change) (reference) = May 1998 reference cash flow 510 × Discount factor of May 1998 + May 1999 reference cash flow 510 × 1999 May discount factor = 2,000 x 99.2 / 100 + 3,000 x 98.8 / 100 Present value of interest rate scenario (before change) (partition) = May 1998 partition cash flow 511 x 1998 May Month discount factor + May 1999 partition cash flow 511 x May 1999 discount factor = 4,000 x 99.2 / 100 + 5,000 x 98.8 / 100 Fixed / fluctuation determination section 120 If the fixed / floating category 502 of the market DB 105 is fluctuating, the subject interest rate scenario ( At Dogo) Shigeyuki creating step 130, already contracted interest rate 30
8. The applicable rate is fixed at the partition rate 309, and there is no need to review the interest rate (part where the interest date has not arrived),
Create cash flows for the subject interest rate scenario (after fluctuations) separately from the portion where the applicable interest rate has not been finalized due to future interest rate renewal (interest date portion).

For example, the reference date 305 = 1997/7
/ 31, deposit date 306 = 1997/5/31, about regular date 307 = 1999/5/31, contracted interest rate 308 = 0.5
%, Partition rate 309 = 0.8%, contracted interest rate review cycle 310 = 12 months, current balance 311 = 1,000,
000 yen, interim payment cycle 313 = 12 months. The contract rate review cycle 310 is 12 months, and the interim payment cycle 313 is 12 months,
The interest and cash flow generated in 1) is the portion where the applicable interest rate is fixed and there is no need to revise the interest rate (the portion where the interest date has not arrived), and the interest and cash flow generated on May 31, 1999 This is the portion (interest date) of which interest cash flows are determined by the subject interest rate scenario (after change).

In the year 503 of the market DB 105, 1998 is 2,0 in the target cash flow 510 in May of that year.
00 is 4,00 in May cash flow 511
0, 1999 in the year 503, 3,000 in the reference cash flow 510 in May of that year, 5,000 in the partition cash flow 511 in May, and May 1998 in the interest rate scenario DB 104. It is assumed that 0.6 is stored in the reference rate and 0.9 is stored in the partition rate.

Further, in interest rate pass generation step 115, V
The discount factor of May 1998 for the interest rate scenario (after fluctuation) created by a method such as the asicek model was 99.2, the discount factor for May 1999 was 98.8, and the interest rate scenario of the item (after fluctuation)
In May 1998, the reference rate is 0.9, and the partition rate is 1.2.

First, in the interest rate scenario (after change) market creation step 130, first, the cash flow of the interest rate due portion of the interest rate scenario (after change) is created as follows.

Cash Flow of Subject Interest Rate Scenario (After Change) (Reference) = Cash Flow of May 1999 Reference Customer 510 × (Category Interest Rate Scenario of May 1998 (Reference) (After Change) / 1998 May interest rate scenario (for review) (before change)) = 3,000 x (0.9 / 0.6) Cash flow (partition) of interest rate scenario (after change) = May 1999 cash partition -Flow 510 x (May 1998 interest rate scenario (partition) (after change) / May 1998 interest rate scenario (partition) (before change)) = 5,000 x (1.2 / 0.9) Then, the interest rate scenario (after fluctuation) market value calculation step 12
At 5, the present value of the interest rate scenario (after fluctuation) is calculated as follows.

Present value of the interest rate scenario (before change) (reference) = May 1998 reference cash flow 510 × discount factor of May 1998 + May 1999 reference cash flow 510 × 1999 May discount factor = 2,000 x 99.2 / 100 + 3,000 x (0.9 / 0.6) x 98.8 / 100 Present value (partition) of interest rate scenario (before change) = 1998 May partition cash flow 511 x discount factor of May 1998 + May 1999 partition cash flow 511 x discount factor of May 1999 = 4,000 x 99.2 / 100 + 5,000 x ( 1.2 / 0.9) × 98.8 / 100 In the simulation number determination step 135, the control program It is determined whether 111 has been repeated for the number of simulations input from the keyboard 101, and if it has been repeated, the present value of the interest rate scenario (after change) and the present value of the interest rate scenario (before change) calculated by repeating the number of simulations are calculated. Using the detailed result calculation step 140
Calculates the expected value and standard deviation of the market price (after fluctuation) in the interest rate scenario, and the market price change width (3 standard deviations, 2 standard deviations, 1.65 standard deviations).

For example, the current evaluation of the interest rate scenario (after fluctuation) in the case of performing the Monte Carlo simulation with three simulations is as follows: the first time = α, the second time = β,
Third time = γ. Further, assuming that the present value of the interest rate scenario (before change) is A, it is calculated as follows.

Interest rate scenario (after change) Expected value of market price m =
(Α + β + γ) / 3 Market value change standard deviation σ = (((α−A) ＾ 2 + (BA))
{2+ (CA)} 2) / 3) ＾ 1/2 Price change (3σ left tail) =-3σ Price change (2σ left tail) =-2σ Price change (1.65σ left tail) = -1.65σ Market price change width (3σ right tail) = 3σ Market price change range (2σ right foot) = 2σ Market price change width (1.65σ right foot) = 1.65σ If the process is repeated for the number of items extracted in step 100, the result of each unit of item calculated as described above is consolidated in the unit of subject and stored in the simulation result DB 107 in the simulation result calculation step 150.

With the above processing, the risk (non-linear risk) that cannot be grasped by the risk metrics is grasped, and the amount of calculation increases exponentially with other calculation methods with respect to the increase in the number of target assets and the like. On the other hand, it increases only in proportion and is efficient.

The second embodiment will be described below.
The difference from the first embodiment is that a commercialization DB (DB for each renewal month) in which the same renewal month is collected for each subject is created in advance, and a simulation is performed using the DB.

FIG. 8 is a flowchart showing a second embodiment of the Monte Carlo simulation to which the present invention is applied.

First, the outline of the flow will be described. First, in a statement extracting step 500, a statement corresponding to the same subject is extracted from the statement DB 103 stored in the storage device.

In the fixed / change determination step 505, the fixed DB or the variable block is determined by the market DB creation program 112 from the detail DB 103 extracted in the detail extraction step 500.

In the case of fixed subjects, in step 510 of making a DB (fixed) month-to-month renewal, the market making DB 112 creates a market DB (DB for month-to-month renewal) based on the information in the detail DB 103. It is stored in a commercialization DB (DB for monthly renewal) 105.

In the case of a variable subject, the contracted interest rate and the partition rate have already been determined in the market DB creation program 112 based on the information in the detail DB 103 in the step of creating a monthly DB (variation 1) for renewal month. Creates a market DB (DB for monthly renewal) 105 for a part that does not need to be reviewed (interest date not yet reached)
105.

Next, in step 520 of creating a monthly DB (variation 2), the information on the detail DB 103 and the subject interest rate scenario DB 104 are input. Market DB for undetermined portion (interest date portion)
B) 105 is created, and the market DB (DB for monthly renewal) 10
5 is stored.

In the interest rate scenario (prior to change) market value calculation step 525, the market interest rate DB (renewal month-by-month DB) 105 and the interest rate scenario DB 106 created above are input and the subject interest rate is calculated by the VAR / EAR calculation program 114. The present value of the scenario (before change) is calculated and stored in the simulation result DB 107.

In the interest rate path generation step 530, the control program 111 sets the interest rate scenario (after the change) or the subject interest rate scenario (after the change) for the number of simulations input from the keyboard 101 as the interest rate pass creation program 1
13 and stored in the work area 115.

In the interest rate scenario (after change) creation step (by renewal month) creation step 535, the market interest rate scenario (after change) created by the market DB (return month by DB) 105 and the interest rate pass generation step 530 are input. , VAR
The / EAR calculation program 114 calculates the execution of the interest rate scenario (after the change) (for each renewal month) and stores it in the work area 115.

In the interest rate scenario (after change) market value calculation step 540, the interest rate scenario (after change) made in the step (535) The VAR / EAR calculation program 114 calculates an item interest rate scenario (after change) market value by using the interest rate scenario (after change) created in the interest rate pass generation step 530 as an input, and stores it in the work area.

Simulation Number Determination Step 545
It is determined whether the control program 111 has repeated the subject interest rate scenario (after change) creation step (by renewal month) 535 and the subject interest rate scenario (after change) market value calculation step 540 for the number of simulations input from the keyboard 101. / EAR calculation program 114
The judgment is made in.

Simulation result calculation step 550
Then, the subject interest rate scenario (after the change) market value calculating step 540 and the subject interest rate scenario (before the change) market value calculating step 540 of the subject interest rate scenario (after the change) obtained by the control program 111 being repeated by the number of simulations input from the keyboard 101. The expected value, standard deviation, and the like are calculated by the VAR / EAR calculation program 114 using the subject interest rate scenario 525 (before fluctuation), and stored in the simulation result DB 107.

First, in the statement extracting step 500, the same subject (ordinary deposit,
The details corresponding to the funding period are extracted based on the account code 304 based on the account code 304 based on the correspondence between the account and the account code defined by the user in advance.

Next, a fixed / floating division determination step 505 is performed.
Then, the market DB creation program 112 determines whether the fixed / variable section 312, which is the extracted information of the detail DB 103, is fixed or variable.

If the extracted details are fixed subjects, the following processing is performed.

In the case of fixed subjects, the contracted interest rate 308,
Since the partitioning rate 309 is fixed and there is no need to review interest rates, step 5 for creating a DB (fixed) for renewal month
At 10, principal cash flow from the base date based on the base date 305, deposit date 306, contract date 307, contracted interest rate 308, partition rate 309, current balance 311, and interim payment cycle 313 of the statement DB 103. , Calculate the cash flow for the customer and the cash flow for the partition.

Then, the calculated principal cash flow, reference cash flow, and partition cash flow of the same item are all added up in the same year and month. Then, the calculated cash flows are stored in the market DB 105. Market DB
The account code corresponding to the account item code is assigned to the account code 501 of the account 105, and the detail DB1 is assigned to the fixed / variable category 502.
03, the fixed / floating category 312, the year 503, the year in which the cash flow occurred, the base date 504, the base date 305 of the detail DB 103, and the base date 504 for the renewal date. Are the principal cash flows 506, 509, 512, 515, and 51 of each month.
8, 521, 524, 527, 530, 533, 53
6, 539, cash flow information on interest 507, 510, 513, which is cash flow information on interest on interest;
516, 519, 522, 525, 528, 531, 5
34, 537, 540, partition interest cash flows 508, 51 which are cash flow information of partition interest.
1, 514, 517, 520, 523, 526, 52
9, 532, 535, 538, and 541 include Step 5
The respective cash flows calculated in step 10 are stored in the market DB 105.

Next, the processing when the extracted subject is a variable subject will be described.

In the case of a variable subject, the applicable interest rate is already determined at the contracted interest rate 308 and the partitioning rate 309 in the step 515 of creating a DB (variation 1) according to the renewal month, and there is no need to review the interest rate (interest rate). Creation of cash flow for the part with no due date) and the cash flow for the part where the applicable interest rate has not been fixed (the part with the due date) in the step 520 of creating the DB (variation 2) for each month Is created by the market creation DB creating program 112 separately from the creation of the database.

First, in step 515 of the DB (variation 1) for monthly renewal of interest, the principal cash flow of the portion where the applicable interest rate has been determined and the interest rate does not need to be reviewed (the portion where the interest date has not yet arrived) is reviewed. Cash flow and partition cash flow are based on the base date 305 and deposit date 30
6, about regular date 307, contract rate 308, partition rate 30
9. Contracted interest rate review cycle 310, current balance 311,
It is calculated from the interim payment cycle 313.

Thereafter, the calculated principal cash flows, reference cash flows, and partition cash flows of the same item are all added up in the same year and month. Then, the calculated cash flows are stored in the market DB 105. In the subject code 501, the subject code of the extracted subject is fixed /
The fluctuation category 502 indicates the fluctuation, the historical year 503 indicates the year in which the cash flow occurs, the reference date 504 indicates the reference date 315 of the detail DB 103, and the renewal date 505 indicates the detail DB 103. The base date of the base date 305 of
Principal cash flows 506, 50 for each month
9, 512, 515, 518, 521, 524, 52
7, 530, 533, 536, 539, and interest-related cash flows 507, 510, 513, 516, 51
9, 522, 525, 528, 531, 534, 53
7, 540, cash flow 508, 5
11, 514, 517, 520, 523, 526, 52
9, 532, 535, 538, and 541 store the calculated values in the market DB.

Next, the principal renewal monthly DB (fluctuation 2) creation program 520 uses the principal cash flow and the customer cash flow for the portion where the renewal is due in the future and the applicable interest rate has not been determined (interest date portion). , The partition cash flow, the base date 305 and the deposit date 31 of the detail DB 103
6, about regular date 307, contract rate review cycle 310,
Current balance 311, interim payment cycle 313, and reference rates 404 and 406 of the interest rate scenario DB 104,
408, 410, 412, 414, 416, 418, 4
20, 422, 424, 426 and partition rates 405, 4
07, 409, 411, 413, 415, 417, 41
9, 421, 423, 425, and 427 are calculated.

Thereafter, the interest rate is reviewed for the same subject in the same year and month, that is, the principal cash flow, reference cash flow, and partition cash flow of the same interest renewal date are all calculated for each period. Add them together. Then, the calculated cash flows are stored in the market DB 105. The subject code 501 indicates the subject code of the extracted subject, the fixed / variable category 502 indicates the variation, the history 503 indicates the date and time when the cash flow occurs, and the reference date 504 indicates the detail DB 103. Of the reference date 305 and the renewal date 505
The deposit date 306 and the date of the interest rate revision obtained from the contracted interest rate revision cycle 310 are defined as the principal cash flows 506, 509, 512, 515 and 51 of each month.
8, 521, 524, 527, 530, 533, 53
6,539, Interest cash flow 507,5
10, 513, 516, 519, 522, 525, 52
8, 531, 534, 537, 540, cash flow 508, 511, 514, 517, 52
0, 523, 526, 529, 532, 535, 53
The calculated values are set in 8,541.

The above-described processing is performed for each subject, and a cash flow for each subject's renewal month is created.

Next, in the subject interest rate scenario (before change) market value calculation step 525, a profit-change month-specific DB (fixed) creation step 510, a profit-change month-specific DB (change 1) creation step 515, and a profit change month-specific DB (change 2) ) The VAR / EAR calculation program 114 inputs the principal, reference and partition cash flows and the discount factor, which is information of the interest rate scenario DB 106, to the market DB 105 created and stored by the creation program 520. The present value (before the change) is calculated and stored in the simulation result DB 107.

Next, in the interest rate pass generation step 530,
Control program 1 from the interest rate pass creation program 113
11 generates an interest rate scenario of the number of simulations input from the keyboard 101 by a method such as a Vasik model, creates an interest rate scenario (after change) or a subject interest rate scenario (after change), and stores it in the work area 115.

Next, in the interest rate scenario (after fluctuating) execution (by renewal month) creation step 535, the interest rate scenario
B (fixed) creation step 510, profit-and-change month-based DB (variation 1) creation step 515, and a sales DB (and / or profit-and-change month-based DB) created by the profit-and-change month-based DB (variation 2) creation program 520, and an interest rate pass generation step 530. Interest rate scenario (after change) and silent interest rate scenario D created in
Using the B104, an interest rate scenario (after the change) is executed (by renewal month).

The portion where the applicable interest rate is fixed and the interest rate does not need to be revised (the portion where the interest rate date has not yet arrived) is the same as the cash flow for each month in which the interest rate scenario has not changed before the interest rate scenario change. Absent.

In the portion where the interest rate is renewed and the applied interest rate is not fixed (the portion where the interest rate is due), the subject interest rate scenario of the subject interest rate scenario (after change) is applied. This is different from the monthly cash flow after the change.

The post-change cash flow for each month of renewal is as shown in the step of creating DB (fixed) for renewal month (step 510), the step of creating DB (variation 1) for renewal month, and the program 520 for creating DB (variation 2) of renewal month. Instead of recreating from the statement DB 103 again, the subject interest rate scenario (before change) corresponding to the renewal date 505 of the market DB (DB for renewal month) 103 and the subject scenario corresponding to the renewal date (after change) Create interest rate scenarios (after change) (by renewal month) at the ratio of.

Thereafter, in the interest rate scenario (after change) in the market value calculation step 540, the interest rate scenario (after change)
The VAR / EAR calculation program 114 inputs the market interest rate scenario (after the change) using the market created by the market creation step 535 (by the month of renewal of interest) and the interest rate scenario (after the change) created by the interest path generation step 530 as inputs. Calculate the current market value of.

Next, in a simulation number determination step 545, the control program 111
Interest rate scenario (after change) for the number of simulations input from
5. Interest rate scenario (after change) market value calculation step 54
The VAR / EAR calculation program 114 determines whether 0 has been repeated.

Finally, in the simulation result calculation step 550, the item code 701 of the simulation result DB 107 contains the item code extracted in the detail extraction step 500, and the customer / partition category 702 holds the customer cash flow. The information as to whether the data is for counting or for the partition cash flow includes the current value of the item interest rate scenario (before change) calculated in the market interest rate scenario (before change) step 525 in the market value before interest rate change 703, The expected value of the market price 704 is the subject interest rate scenario (after change) market price calculated in the step 540 for calculating the subject interest rate scenario, and the market price change standard deviation 705 is the subject interest rate scenario (before change) the current price calculation step. 52
5. Interest rate scenario (after change) market value calculation step 54
The standard deviation is calculated by using the value calculated at
6, 707, 708, 709, 710, and 711 store the market price change width using the values calculated in the subject interest rate scenario (before change) market value calculation step 525 and the subject interest rate scenario (after change) market price calculation step 540. I do.

Hereinafter, a specific example will be described.

The detail extracting step 500 allows the detail DB1
03, the fixed / variable division 312 of the item is fixed, the base date 305 = 1997/7/31, the deposit date 3
06 = 1997/5/31, about regular date 307 = 1999
/ 5/31, contracted interest rate 308 = 0.5%, partition rate 3
09 = 0.8%, current balance 311 = 1,000,000
Yen, interim payment cycle 313 = 12 months.

Profit Revision Monthly DB (Fixed) Creation Step 51
At 0, the interim payment cycle 313 is 12 months, and the deposit date 306 is the value of the reference and partition interest cash flows generated on May 31, 1998 and May 31, 1999 from May 31, 1997.

The value of the reference and partition cash flows generated on May 31, 1998 is calculated as follows:
Since the interest cash flow is up to 1, the interest cash flow for the customer = the current balance 311 x the contracted interest rate 308 x (the number of days from the base date 305 to the date of interest generation) / 365 = 1,000,000 x 0.5 / 100 x (days from July 31, 1997 to May 31, 1998) / 3 365 partition interest cash flow = current balance 311 x partition rate 309 x (days from base date 305 to date of interest occurrence) /365=1,000,000×0.8/100×(days from 1997/7/31 to 1998/5/31) / 365.

The value of the reference and partition cash flows generated on May 31, 1999 is based on the previous interim interest payment date of 19th.
Next interim payment date from May 31, 1998 May 3, 1999
Since the interest cash flow is up to 1, interest cash flow to the customer = current balance 311 x contracted interest rate 308 x (days from the previous interim interest payment date to the next interim interest payment date) / 365 = 1,000,000 x 0. 5/100 x (days from May 31, 1998 to May 31, 1999) / 3 65 Partition interest cash flow = current balance 311 x partition rate 308 x (from the previous interim payment date to the next interim payment date) Days) /365=1,000,000×0.8/100× (days from 1998/5/31 to 1999/5/31) / 365. In addition, about 307 on 1999/5
/ 31 has a principal cash flow of 1,000,000
0 also occurs.

Thereafter, the cash flow generated in the same year and month in the principal cash flow, the reference cash flow, and the partition cash flow of the item of the same item calculated in step 510 of creating a DB (fixed) for renewal month. Add all the flows.

The cash flow information obtained as described above is stored in the market DB 105.

The detail extraction step 500 allows the detail DB1
If the fixed / floating category 312 of the details extracted from the section 03 is fluctuating, the applicable interest rate has already been determined at the contracted interest rate 308 and the partition rate 309, and there is no need to review the interest rate (part where the interest date has not arrived). There is a portion where the interest rate is due to change in the future and the applied interest rate is not fixed (interest date portion).

First, in step 515 of creating a DB (variation 1) for monthly renewal of interest, the principal cash flow of the portion where the applied interest rate has been determined and the interest rate does not need to be reviewed (the portion where the interest date has not yet arrived) is calculated. The customer cash flow and the partition cash flow are calculated based on the base date 305, deposit date 306, contract date 307, contract rate 308, partition rate 309, contract rate review cycle 310, and current balance 31 of the transaction statement.
1. Calculate from the interim payment cycle 313.

For example, reference date 305 = 1997/7
/ 31, deposit date 306 = 1997/5/31, about regular date 307 = 1999/5/31, contracted interest rate 308 = 0.5
%, Partition rate 309 = 0.8%, contracted interest rate review cycle 310 = 12 months, current balance 311 = 1,000,
000 yen, interim payment cycle 313 = 12 months. The contract rate review cycle 310 is 12 months, and the interim payment cycle 313 is 12 months,
The reference and interest cash flows generated in (1) are fixed and calculated as follows.

The value of the reference and partition cash flows generated on May 31, 1998 is 1
Interim payment date from 997/7/31 1998/5/31
Since the interest cash flow until then, the interest cash flow for the customer = current balance 311 x contracted interest rate 308 x (days from the base date 305 to the date of interest generation) / 365 = 1,000,000 x 0.5 / 100 x (days from July 7, 1997 to May 31, 1998) / 3 65 Partition interest cash flow = current balance 311 x partition rate 309 x (days from base date 305 to date of interest) / 365 = 1,000,000 x 0.8 / 100 x (days from 1997/7/31 to 1998/5/31) / 3365.

Next, in step 520 of creating a DB (variation 2) for recurring monthly renewal, the principal cash flow and the customer cash flow of the portion where the renewal is due in the future and the applicable interest rate is not fixed (the part where the interest date expires) , The partitioning cash flow, the base date 305 of the transaction statement, the deposit date 316, the regular date 307, the contracted interest rate review cycle 310, the current balance 311, the intermediate interest payment cycle 313, and the storage device 10
8, the reference rate 404 of the interest rate scenario DB 104,
406, 408, 410, 412, 414, 416, 4
18, 420, 422, 424, 426 and partition rate 4
05, 407, 409, 411, 413, 415, 41
7, 419, 421, 423, 425, and 427 are calculated.

For example, in the same manner as described above, in the case of a variable period subject, the base date 306 = 1997/7/31, the deposit date 306 = 1997/5/31, and the approximate period 307 = 19
99/5/31, contracted interest rate review cycle 310 = 12
Month, current balance 311 = 1,000,000 yen, interim payment cycle = 12 months. The reference and cash interest cash flows generated on May 31, 1999 are calculated based on the reference rates 404, 406, 408, 410 of the interest rate scenario DB 104 of 1998/5, in which the contracted interest rate review cycle 310 is in the future from 12 months. , 412, 41
4, 416, 418, 420, 422, 424, 426
And partition rates 405, 407, 409, 411, 4
13, 415, 417, 419, 421, 423, 42
5, 427 are calculated. Assume that the reference rate of 1998/5 = 0.6 and the partition rate = 0.9. 1999/5 /
31 The value of the reference and partition cash flows generated is the interest cash flow from the previous interim payment date of 1998/5/31 to the next interim payment date of 1999/5/31. 311 x reference rate x (the number of days from the previous interim payment date to the next interim payment date) / 365 = 1,000,000 x 0.6 / 100 x (from 1998/5/31 to 1999/5/31 Days) / 3 65 partition interest cash flow = current balance 311 x partition rate x (days from last interim payment date to next interim payment date) / 365 = 1,000,000 x 0.9 / 100 x (1998) / 5/31 to 1999/5/31) / 365. In addition, about 307 on 1999/5
/ 31 has a principal cash flow of 1,000,000
0 also occurs.

Thereafter, the interest rate is reviewed for the same year and month in the same subject calculated by the profit-change month DB (variation 1) step 515 and the profit-change month DB (variation 2) creation program 520, that is, the profit change date is the same. The value of this cash flow, the value of the customer cash flow, and the value of the partition cash flow are all added up for each period.

The cash flow information obtained as described above is stored in the market DB 105.

In the year 503 of the market DB 105, the year 1998 is changed to 2,0 by the target cash flow 510 in May of that year.
00 is 4,00 in May cash flow 511
Suppose that 0 is stored, 3,000 is stored in the reference cash flow 510 in May, 1999, and 5,000 is stored in the partition cash flow 511 in May in the year 503 in the year.

In the interest rate scenario DB 106, the year 60
1 to 1998 is 99.5 for the May discount factor 603 for the year, 1999 to 991 for the May discount factor 603 for the year 601.
It is assumed that 0 is stored.

Then, in the interest rate scenario (before change) market price calculation step 525, the interest rate scenario (before change)
Is calculated as follows.

Present value of interest rate scenario (before change) (reference) = Cash flow 510 in fiscal year 503 in May and reference to cash flow 510 x discount factor in fiscal year 503 in May and 603 + fiscal year 503 in 1999 May reference cash flow 510 x year 503 is 1999 and May discount factor 603 = 2,000 x 99.5 / 100 + 3,000 x 99.0 / 100 Present value of interest rate scenario (before change) (partition) ) = Cash flow 511 for fiscal 503 in May and May 511 × Discount factor for fiscal 503 in May and May 603 + Cash flow for fiscal 503 in May and May 511 × Fiscal 503 for May and 1999 Discount factor 603 = 4,000 x 99.5 / 10 + 5,000 × 99.0 / 100 then calculates the rates scenario (after change) or subjects rates scenario (after change) by a technique such as Vasicek model.

In the interest rate scenario (after change) market (change by month) step 535, a DB (fixed) change month 510, a DB by change month (change 1) step 515, and a DB by change month (change) 2) Market DB created by the creation program 520 (DB for monthly renewal)
105 and the subject interest rate scenario (after change) created in the interest rate pass generation step 530 and the subject interest rate scenario DB 104
, Create an interest rate scenario (after change) and execute (by renewal month). The portion where the applicable interest rate is fixed and the interest rate does not need to be reviewed (the portion where the interest date has not yet arrived) is
Since the renewal date 505 of the B (renewal month DB) 105 is the same as the reference date 305 of the detail DB 103, the market DB (renewal month DB) 105 is not modified.

The portion where the applicable interest rate has not been settled in the future when the interest rate is renewed (interest date portion) is recorded in the market DB (D
B) Execution of the subject interest rate scenario (after change) by the ratio of the subject interest rate scenario (before change) corresponding to the renewal date 105 of 105 and the subject interest rate scenario corresponding to the renewal date 505 (after change). (Renewal month by month) is created in the same manner as the subject interest rate scenario (after change) market creation step 130.

Thereafter, in the interest rate scenario (after the change) in the market value calculation step 540, the interest rate scenario (after the change)
Is calculated in the same manner as in the interest value scenario (after fluctuation) market value calculation step 125.

Simulation number determination step 135
It is determined whether the control program 111 has repeated the number of simulations input from the keyboard 101, and if so, the present value of the subject interest rate scenario (after fluctuation) and the subject interest rate scenario (after the change) calculated by repeating the number of simulations In the simulation result calculation step 550, the expected value and standard deviation of the market interest rate scenario (after the fluctuation) and the fluctuation range of the market price (3 standard deviations, 2 standard deviations, 1.65 standard deviations) using the present value of the ) Is calculated similarly to the detailed result calculation step 140. After that, it is stored in the simulation result DB 107.

As described above, by creating in advance a market DB (DB for each month of renewal of interest) 105 that aggregates the same interest renewal month for each subject, it is possible to maintain calculation accuracy and This is a risk calculation method that is tens of times and hundreds of times faster than the risk calculation process performed in the above.

Although the embodiment of the risk calculation (value-at-risk) has been described, the risk index (earning-at-risk) can be calculated by the same processing. In the flowchart of FIG. 2, the fixed / change determination step 120 and the subject interest rate scenario (after change) market creation step 130 are not performed. In the flowchart of FIG. 8, the interest rate scenario (after fluctuation) is executed (by month of renewal of interest).
This is a process in which the creation step 535 is not performed.

FIG. 9 shows an example of a display screen 900 for displaying the risk index calculated as described above on the display 100.

A risk calculation result of a certain subject is displayed in a table 901 and a bar graph 902. The market price change width in the table 901 is the market price change width (market price after interest rate fluctuation-market price before interest rate fluctuation) obtained from the simulation result. Based on this, a value is set by equally dividing the maximum and minimum values into the number of simulation divisions. The frequency sets the number of simulations corresponding to the market price change width.
The probability is calculated by the control program 111 from the number of simulations and the frequency input from the keyboard 101. A bar graph 902 is a bar graph of the table 901.

Table 903 is a table in which the values obtained in the simulation result calculation step 150 are set.

The subject selection button 950 is displayed on the screen 900
The setting of the subject to be displayed can be switched.

The simulation re-execute button 955 is
The simulation can be performed again based on the result of the calculated risk index. Simulation button 9
When the button 55 is pressed, the display screen 1000 shown in FIG.

The risk calculation results of a certain subject are displayed in Tables 1001 and 1002. The market price change width in Table 1001 is the market price change width (market price after interest rate fluctuation minus market price before interest rate fluctuation) obtained from the simulation result. Based on this, a value is set by equally dividing the maximum and minimum values into the number of simulation divisions. The frequency sets the number of simulations corresponding to the market price change width.
The maximum number of times number sets the number of simulations that achieved the largest market value change width among the market price change ranges, and the maximum value market value change range sets the value at that time. Similarly to the maximum number of times, the minimum number of times is the number of simulations that achieves the minimum market value of the range of changes in market price, and the minimum value of the market value is
Set the value at that time.

Table 1002 is a table in which the values obtained in the calculation result calculation step 150 are set.

The subject selection button 1050 is displayed on the screen display 10
Switching of the setting of the subject displayed at 00 is enabled.

Simulation re-execute button 1055
Enables the simulation to be performed again based on the calculated result of the risk index. Select the maximum value frequency number and the minimum value frequency number displayed in Table 1001,
When the simulation re-execute button 1055 is pressed,
The simulation at the time of realizing it can be executed again.

Also, an interest rate scenario reoccurrence button 1060
Enables an interest rate scenario to be generated again based on the result of the calculated risk index. When the maximum number of times number and the minimum number of times number displayed in the table 1001 are selected and the interest rate scenario reoccurrence button 1060 is pressed,
The interest rate scenario when this is realized can be generated again.

The previous screen button 1065 is displayed on the display screen 900.
Button for returning to.

As described above, according to the present invention, it is possible to improve the processing speed while maintaining the accuracy.
Simulation can be realized.

[0163]

As described above, according to the present invention, the value-at-risk and the learning-at-risk can be calculated at a high speed by performing the calculation by the Monte Carlo simulation.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing an example of a risk calculation device applied to the present invention.

FIG. 2 is a flowchart showing one embodiment of a risk calculation using detailed information in the processing procedure of the present invention.

FIG. 3 is a data configuration diagram showing an example of information stored in a statement DB.

FIG. 4 is a data configuration diagram showing an example of information stored in a subject interest rate scenario DB.

FIG. 5 is a data configuration diagram showing an example of information stored in a market DB.

FIG. 6 is a data configuration diagram showing an example of information stored in an interest rate scenario DB.

FIG. 7 is a data configuration diagram showing an example of information stored in a simulation result DB.

FIG. 8 is a processing procedure of the present invention;
6 is a flowchart illustrating an example of risk calculation using B).

FIG. 9 is a diagram showing a first example of a screen display of a risk calculation result on the display screen of the present invention.

FIG. 10 is a diagram showing a second example of a screen display of a risk calculation result on the display screen of the present invention.

[Explanation of symbols]

100 ... display, 101 ... keyboard, 102 ...
CPU, 103: detail DB, 104: subject interest rate scenario DB, 105: market DB 106: interest scenario DB, 107: simulation result DB, 1
08: storage device 109: bus, 110: memory, 111: control program, 112: commercial DB creation program, 113: interest rate pass creation program 114: VAR / EAR calculation program, 115 ...
Work area 120: CD-ROM driver.

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Kenji Ando 890 Kashimada, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture Information Systems Division, Hitachi, Ltd. (72) Inventor Takao Sekimoto 12 Hitachi, Ltd. Information Systems Division (72) Inventor Takanobu Hanaoka 890 Kashimada, Saiwai-ku, Kawasaki-shi, Kanagawa 12 Hitachi Information Systems Division (72) Inventor Nobuaki Saito Hara 2-chome F-term in Hitachi, Ltd. System Development Division (Reference) 5B049 AA00 AA04 BB46 CC00 EE41 5B056 AA00 AA04 BB00 BB62 BB64

Claims (1)

[Claims] Claims 1. A method of calculating risk using transaction information of financial instruments and a scenario of a future interest rate for each subject,
A risk calculation method characterized by calculating a value-at-risk or an learning-at-risk based on detailed information of a certain subject and a subject interest rate scenario.