JP2005275837A - Device and program for managing risk assessment index value data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a risk assessment index value data management device for performing management with excellent reproducibility and a fewer amount of data concerning a large amount of risk assessment index value data which is the aggregation of discrete index values generated by a Monte Carlo simulation to assess a business risk. <P>SOLUTION: The device comprises: a data input part 11 for receiving the input of the risk assessment index value data and temporarily preserving it; a setting value storage part 12 for storing the discrete percentile setting value of the generation probability of the risk assessment index value data which are inputted to the data input part; a data preserving part 15 for extracting a percentile value, the maximum value, and the minimum value in the percentile setting value of a risk assessment index value from the temporarily preserved risk assessment index value data and preserving them in a risk assessment index value database 14; and a data reproducing part 16 for extracting the respective percentile value, the maximum value, and the minimum value of the risk assessment index value from the risk assessment index value database 14, calculating a probability density in each percentile setting value, and reproducing a probability density distribution. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a risk evaluation index value data management apparatus for managing risk evaluation index value data, which is a set of discrete index values generated by Monte Carlo simulation for evaluating business risk, on a computer, and more specifically. The present invention relates to a risk evaluation index value data management device for managing risk evaluation index value data in a risk management system for managing each business risk for each portfolio item.

  For example, when installing equipment to a customer company, the customer does not purchase the equipment, but introduces it at the expense of the equipment seller (initial investment), and the customer company who is the user of the equipment In some cases, the equipment is introduced by collecting a usage fee over a long period of time and collecting the initial investment. In the case where the equipment is a large-scale cogeneration system for business establishments, it is assumed that the customer company uses the electric power and heat energy generated by the system and collects the usage fee. When the investment is made to multiple client companies, the equipment distributor analyzes the risk and return for each investment project, and performs risk management using the so-called portfolio theory for the multiple projects. Do. The risk management is executed using a risk management system that uses software processing (simulation) of a computer.

  Here, as an index used for the analysis of risk and return described above, investment is made based on the cash flow (CF), which is one of the financial indicators for each fiscal year for each project, and the total present value of each cash flow for each fiscal year. NPV (net present value: net present value) given by a value obtained by subtracting the amount, VaR (value at risk), which is one of risk management indexes, and the like are calculated using the risk management system. In addition, about each parameter | index (NPV, VaR, etc.) used for risk management, the following patent document 1 and 2 etc. have description.

  Assuming multiple scenarios (for example, the number of scenarios of about 5000 to 10,000), the Monte Carlo simulation is executed for each scenario, and a large amount of cash flows distributed with a certain probability of occurrence. Cash flow data is obtained, and predetermined risk and return analysis is performed based on the data.

By the way, the index values for risk evaluation such as cash flow for each fiscal year along the time series for each project generated by the above Monte Carlo simulation are stored for later use in risk evaluation of the entire portfolio. It is necessary to keep it. In general, a large amount of data subject to statistical processing is conventionally 1) stored as it is, 2) stored as a histogram, or 3) applied to standard known probability distributions such as normal distribution and lognormal distribution. The method of storing it was taken.
JP 2003-296558 A JP 2003-108756 A

  However, risk assessment index values such as cash flow for each fiscal year along the time series for each case generated by the Monte Carlo simulation are generated for multiple scenarios for each fiscal year, for example 10 to 20 years ahead, If the number of items to be simulated exceeds 100 and the number of scenarios is about 10,000 or more, the amount of data becomes extremely large. For example, risk management processing cannot be easily performed using a personal computer or the like, which is unrealistic. In addition, when a large amount of data is stored as a histogram, there is a problem in reproducibility. In other words, since the width of the histogram is constant, when trying to reproduce as a probability distribution, it is impossible to precisely reproduce only the skirt part with a small probability density. It is necessary to make the width of the histogram fine, and the amount of data increases. In addition, the calculation of the histogram requires a calculation process for that purpose, which increases the labor of the process. Furthermore, by applying the data generated by the Monte Carlo simulation to a known probability distribution such as a normal distribution, the characteristics of the data obtained by executing a complicated simulation will be eliminated in the first place. There is a problem.

  The present invention has been made in view of the above-described problems, and an object thereof is a set of discrete index values generated by Monte Carlo simulation for solving the above problems and evaluating business risk. An object of the present invention is to provide a risk evaluation index value data management device and a risk management system capable of managing a large amount of risk evaluation index value data with high reproducibility and a small data amount.

  In order to achieve the above object, a risk evaluation index value data management apparatus according to the present invention stores risk evaluation index value data, which is a set of discrete index values generated by Monte Carlo simulation for evaluating business risk, on a computer. A risk evaluation index value data management apparatus that manages the risk evaluation index value data, accepts input of the risk evaluation index value data, and temporarily stores it in a predetermined storage area; and the risk evaluation input to the data input section A setting value storage unit that stores discrete percentile setting values of occurrence probability of index value data; and the risk evaluation that is the temporary storage of the percentile value, maximum value, and minimum value in the percentile setting value of the risk evaluation index value A data storage unit that is extracted from the index value data and stored in the risk evaluation index value database; Ete becomes possible to first characterized.

  Furthermore, in addition to the first feature, the risk evaluation index value data management apparatus extracts the percentile value, the maximum value, and the minimum value of the risk evaluation index value from the risk evaluation index value database, It comprises a data reproduction unit that calculates the probability density in each percentile setting value based on each extracted percentile value, the maximum value, and the minimum value, and reproduces the probability density distribution of the risk evaluation index value. The second feature.

According to the risk evaluation index value data management device of the first or second feature, the probability density distribution obtained approximately by the discrete frequency distribution of a large amount of risk evaluation index value data generated by the Monte Carlo simulation is accurately obtained. The data amount of the risk evaluation index value data can be compressed and stored so that it can be reproduced well. That is, by storing the percentile value, the maximum value, and the minimum value in the preset percentile setting value of the risk evaluation index value, the data amount can be greatly reduced. Further, if the percentile value X ₁ (risk evaluation index value) at the minimum value (X ₀ ) and the minimum percentile setting value (x ₁ %) is set, the minimum value X ₀ and the risk evaluation index value X ₁ of the probability density distribution curve since the probability generating between (corresponding to the area) corresponds to percentile setpoint x _1%, the probability density p (x ₁₎ in percentile setpoint x _1%, from the minimum value X ₀ and risk evaluation index value X ₁ It can be calculated approximately. Similarly, the minimum percentile setpoint percentile value at _{(x 1%) X} 1 and the following percentile setting from percentile _{X 2} in _(x 2%), the probability density p in percentile setpoint _{x 2% (x 2)} Can be calculated. Since the probability density p (x _i ) at the upper percentile setting value (x _i %) can be calculated sequentially, the maximum value and minimum value (each probability density is 0) and the probability at each percentile setting value (x _i %) If the density p (x _i ) is connected by a broken line, the original probability density distribution curve can be reproduced by a broken line approximation. The percentile setting values are set in the range of 0 to 100%, and each setting value represents the occurrence probability (%) of the risk evaluation index value below the percentile value in the setting value. Therefore, the occurrence probability of the risk evaluation index value between the percentile value (x _m %) and the percentile value (x _n %) is (x _n −x _m )%. Specifically, the area of the trapezoid or triangle surrounded by the horizontal axis and the vertical line drawn from the both sides of each broken line of the probability density distribution curve of the broken line approximation with the risk evaluation index value on the horizontal axis and the probability density on the vertical axis Is expressed as a probability of occurrence of a risk evaluation index value between percentile values at both ends of each broken line, and a probability distribution is obtained. Here, by approximating the probability density distribution curve with a broken line, the tail portion having a small probability density can be reproduced with higher accuracy than the approximation by the histogram, which is convenient in risk analysis.

  In particular, according to the risk evaluation index value data management apparatus of the second feature, the percentile value, the maximum value, and the minimum value of the risk evaluation index value stored in the risk evaluation index value database by the data storage unit by the data reproduction unit. Thus, the probability density distribution of the original risk evaluation index value can be accurately reproduced.

  Further, the risk evaluation index value data management apparatus is a risk evaluation index value data for managing risk evaluation index value data, which is a set of discrete index values generated by Monte Carlo simulation for evaluating business risk, on a computer. A management device that accepts input of the risk evaluation index value data and temporarily stores it in a predetermined storage area; and a discrete occurrence probability of the risk evaluation index value data input to the data input section A set value storage unit that stores typical percentile setting values, and extracts the percentile value, maximum value, and minimum value in the percentile setting values of the risk evaluation index value from the temporarily stored risk evaluation index value data, and extracts Each percentile setting is based on the percentile value, the maximum value, and the minimum value. A probability density calculation unit for calculating a probability density in a value; a data storage unit for storing each probability density, the maximum value, and the minimum value in the percentile setting value of the risk evaluation index value in a risk evaluation index value database; The third feature is that the above-described features are provided.

  Further, in addition to the third feature, the risk evaluation index value data management device further includes the probability density, the maximum value, and the maximum value in the percentile setting value of the risk evaluation index value from the risk evaluation index value database. According to a fourth feature of the present invention, a data reproduction unit that extracts a minimum value and reproduces a probability density distribution of the risk evaluation index value based on the extracted probability density, the maximum value, and the minimum value is provided. To do.

  According to the risk evaluation index value data management device of the third or fourth feature, the probability density distribution obtained approximately by the discrete frequency distribution of a large amount of risk evaluation index value data generated by the Monte Carlo simulation is accurate. The data amount of the risk evaluation index value data can be compressed and stored so that it can be reproduced well. That is, by storing the probability density, the maximum value, and the minimum value in the preset percentile setting value of the risk evaluation index value, the data amount can be greatly reduced. Similarly to the first or second feature, the probability density, maximum value, and minimum value of each percentile setting value stored in the risk evaluation index value database are accurately reproduced by a polygonal line approximation of the original probability density distribution curve. it can. In the third or fourth feature, since the probability density is calculated at the time of storage, it is not necessary to calculate the probability density at the time of reproduction. Since the probability density at each percentile set value is stored, it is not necessary to store the percentile value (risk evaluation index value) at each percentile set value, and it is possible to back-calculate from the probability density as necessary.

  In particular, according to the risk evaluation index value data management apparatus of the fourth feature, the probability density, maximum value, and minimum value of the risk evaluation index value stored in the risk evaluation index value database by the data storage unit by the data reproduction unit Thus, the probability density distribution of the original risk evaluation index value can be accurately reproduced.

  Further, in the risk evaluation index value data management device, in addition to any of the above features, the percentile setting value is set to be subdivided so as to increase toward the bottom of the probability density distribution with a small probability density. Is the fifth feature.

  According to the risk evaluation index value data management device of the fifth feature, the tail portion having a small probability density can be reproduced more accurately, which is convenient in risk analysis.

  Furthermore, in addition to any of the above features, the risk evaluation index value data management device further includes a setting value input unit that receives an input of the percentile setting value and stores it in the setting value storage unit. It is characterized by.

  According to the risk evaluation index value data management device of the sixth feature, the user can freely change the setting of the percentile set value. As a result, for example, an appropriate percentile setting value can be set for each case.

  In order to achieve the above object, the risk management system according to the present invention includes a risk evaluation index value data management apparatus having any of the above characteristics, risk and return analysis for each portfolio item, and business risk analysis for the entire portfolio. Characterized in that it comprises: a portfolio analysis device that executes software processing of a computer using a predetermined analysis index; and a simulation device that generates the risk evaluation index value by Monte Carlo simulation based on a plurality of scenarios. .

  According to the risk management system of the present invention, a large amount of risk evaluation index value data can be managed with high reproducibility and with a small amount of data. Therefore, the risk for each portfolio item can be accurately determined using simple hardware resources. And return analysis, and business risk analysis of the entire portfolio.

  In order to achieve the above object, a risk evaluation index value data management program according to the present invention includes the data input unit, the data storage unit, and the data in the risk evaluation index value data management device having the second feature. A first feature is that the computer has a computer program for executing each process of the reproduction unit on the computer.

  According to the risk evaluation index value data management program of the first feature, the risk evaluation index according to the present invention that exhibits the operational effects of the first and second features by executing each computer program on a computer, respectively. A value data management apparatus is realized on the computer.

  Furthermore, the risk evaluation index value data management program includes the data input unit, the probability density calculation unit, the data storage unit, and the data reproduction unit in the risk evaluation index value data management device having the fourth feature. A second feature is that a computer program for executing each of the above processes on the computer is provided.

  According to the risk evaluation index value data management program of the second characteristic, the risk evaluation index according to the present invention that exhibits the operational effects of the third and fourth characteristics by executing each computer program on a computer, respectively. A value data management apparatus is realized on the computer.

  Embodiments of a risk evaluation index value data management apparatus (hereinafter referred to as “present invention apparatus”) and a risk management system (hereinafter referred to as “present invention system”) according to the present invention will be described with reference to the drawings. To do.

  In the present embodiment, as shown in FIG. 1 or FIG. 6, the device 1 of the present invention uses a risk and return analysis for each item of a portfolio including a prospective item, a business risk analysis for the entire portfolio, and the like. As an index, cash flow, which is one of the financial indicators for each fiscal year for each project, NPV given by subtracting the amount of investment from the total present value of each cash flow for each fiscal year, and risk management index 1 One is the portfolio analysis device 2 that is executed by computer software processing using VaR, etc., and the cash flow for each year for each case (corresponding to the index value “risk evaluation index value” for evaluating business risk) Together with the simulation device 3 that generates Monte Carlo simulations based on a plurality of scenarios. The risk management index value data, which is a set of risk evaluation index values (cash flow) generated by Monte Carlo simulation, is configured on a predetermined computer 5 as a part of the network management system (system of the present invention) 4 A software processing device managed on the computer 5.

<First Embodiment>
As shown in FIG. 1, the device 1 of the present invention accepts input of risk evaluation index value data generated by the simulation device 3 and temporarily stores it in a predetermined storage area. A setting value storage unit 12 that stores discrete percentile setting values of the probability of occurrence of the input risk evaluation index value data; a setting value input unit 13 that receives input of percentile setting values and stores them in the setting value storage unit 12; A data storage unit 15 that extracts the percentile value, maximum value, and minimum value of each percentile setting value of the risk evaluation index value from the temporarily stored risk evaluation index value data, and stores the extracted data in the risk evaluation index value database 14; Each percentile value, maximum value, and minimum value of the risk evaluation index value are extracted from the index value database 14, and each extracted parameter is extracted. The probability density calculated at each percentile set value based on Sentairu and maximum and minimum values, and a data reproduction unit 16 to reproduce the probability density distribution of the risk evaluation index value. Here, the data input unit 11, the set value input unit 13, the data storage unit 15, and the data reproduction unit 16 are realized by executing a program for processing each unit separately on the computer 5. Further, the set value storage unit 12 and the risk evaluation index value database 14 are respectively constructed in a predetermined storage area of a predetermined storage device provided in the computer 5. The set value storage unit 12 and the risk evaluation index value database 14 may be configured in separate storage devices.

  Next, the operation procedure of the device 1 of the present invention and the system 4 of the present invention will be described with reference to the flowcharts shown in FIGS.

  First, in the risk and return analysis processing for each individual case, the portfolio analysis device 2 of the system 4 of the present invention uses a user input device (keyboard, pointing device) on a predetermined input screen displayed on the display device of the computer 5. Etc.) is selected and an individual case is selected (# 1). Next, an input of a condition for scenario setting by a similar input operation is received, and a plurality of scenarios (for example, the number of scenarios is about 5000 to 10000) is generated (# 2).

  Next, the simulation apparatus 3 of the system 4 of the present invention executes a Monte Carlo simulation based on a plurality of scenarios generated in Step # 2 for the case designated in Step # 1, for example, in the first year to The cash flow (an example of the risk evaluation index value) at the future time point until the 20th year is calculated (# 3).

  Next, the portfolio analysis apparatus 2 accepts the input of the discount rate (interest rate) by the user's input operation, and from the frequency distribution (or probability density distribution) of the cash flow, NPV (net present current) as expected profit (return) Value) frequency distribution (or probability density distribution) or probability distribution (cumulative value), and, for example, 95% VaR value (value that can be expected to be more profitable with 95% reliability probability) as business risk (# 4) The calculation result is displayed on the screen of the display device of the computer 5 as a graph, a table or the like (# 5). Thereby, the user can perform business feasibility evaluation for individual projects.

  Next, the device 1 of the present invention sets a thinning interval for thinning out the number of data of the risk evaluation index value data generated in step # 3 and storing it in the risk evaluation index value database 14 (# 6). This thinning interval is set as a percentile setting value. The percentiles are, for example, 0%, 3%, 5%, 10%, 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 85%, 90%, 95%, It is set to 100%. The percentile values (risk evaluation index values) at the percentile setting values 0% and 100% correspond to the minimum value and the maximum value of the risk evaluation index value. In the present embodiment, each percentile set value is received by the set value input unit 13 in advance by the set value input unit 13, and the data storage unit 15 stores the set value stored in the set value storage unit 12. Read and use. In addition, each percentile setting value is set such that a portion with a small probability density (bottom portion) in the probability density distribution is subdivided from other portions. As an example of setting the percentile setting value, in addition to the above, for example, 0% to 10% and 90% to 100% are incremented by 1%, 10% to 20% and 80% to 90% are incremented by 2%, and the others are 5 % Increments etc. are acceptable.

  Next, the data storage unit 15 of the device 1 of the present invention, as shown in FIG. 4, the percentile value (risk evaluation index value) in each of the percentile setting values in the risk evaluation index value data generated in step # 3. Are extracted and stored in the risk evaluation index value database 14 (# 7). Naturally, the minimum and maximum values at the percentile setting values 0% and 100% are also stored. At the same time, the average value and the variance calculated based on the original data before thinning out the data are simultaneously saved. Here, since the risk evaluation index value data is generated every year, the thinning-out storage is also executed independently every year. Furthermore, only the percentile value (NPV) at the same percentile setting value is stored for the NPV (net present value) generated in step # 4. In addition, the 95% VaR value of each year generated in step # 4 is also stored.

  Here, it is determined whether or not the individual cases for performing the processes of steps # 1 to # 7 remain (# 8). If other cases remain, the processes of steps # 1 to # 7 are performed. Execute repeatedly. In addition, when the processing of steps # 1 to # 7 has been completed for all cases, a confirmation message asking whether or not to analyze the business risk in the entire portfolio is displayed on the screen of the display device of the computer 5 Output above and accept the response input of the user (# 9).

  When a confirmation input for analyzing business risk in the entire portfolio is accepted in step # 9, the process proceeds to business risk analysis processing in the entire portfolio shown in the flowchart of FIG.

  When the business risk analysis process for the entire portfolio is started, the data reproduction unit 16 of the device 1 of the present invention accesses the risk evaluation index value database 14 and the risk evaluation index for each percentile setting value for each year of all projects. The value (cash flow) and NPV are extracted (# 10).

Subsequently, the data reproduction unit 16 reproduces the probability density distribution shape from the extracted risk evaluation index value (# 11). Specifically, each percentile setting value is x _i (i = 0, 1, 2, 3,..., N), the percentile value in each percentile setting value x _i is X _i , and the probability density is p (x _{If i} ), the probability density p (x _i ) is given by the following equation (1). However, when x ₀ = 0 (%) and x _n = 100 (%), X ₀ and X _n are the minimum value and the maximum value of the risk evaluation index value, respectively. Further, p (x ₀ ) = p (x _n ) = 0.

(Equation 1)
p (x ₁ ) = 0.02 × x ₁ / (X ₁ −X ₀ )
p (x _i ) = 0.02 × (x _i −x _i−1 ) / (X _i −X _i−1 ) −p (x _i−1 )

When the probability density p (x _i ) (i = 0 to n) at each percentile setting value x _i calculated by the recurrence formula of the first formula and the second formula of Formula 1 is connected by a broken line, as shown in FIG. In addition, the probability density distribution of the original risk evaluation index value is reproduced by broken line approximation. Here, the area of the trapezoid between the percentile values X _i and X _i−1 of any two adjacent percentile setting values x _i and x _i−1 is the risk evaluation index value is the percentile value X _i and X _{i−. Since} the occurrence probability (x _i -x _i-1 ) occurs during 1, the recursion formula of the second formula of Formula 1 is derived. Since the percentile values X ₀ and X ₁ of the percentile setting values x ₀ and x ₁ are triangular, the first expression of Equation 1 is derived.

  Next, the portfolio analysis apparatus 2 calculates the expected return and return variance (the risk index because of the fluctuation of the return), the 95% VaR value, etc. The calculation result is displayed on the screen of the display device of the computer 5 (# 12). Further, as the risk index, the square root (standard deviation) of the variance may be used. Here, the above calculation procedure is briefly explained. For example, assuming that there are two cases A and B, the return (revenue or rate of return) of the entire portfolio is R, and the individual returns of cases A and B are X , Y and the investment allocation is a and b (where a + b = 1), the return R for the entire portfolio is expressed by Equation 2, the expected return E (R) and the variance Var (R) for the entire portfolio Are given by Equation 3 and Equation 4, respectively.

(Equation 2)
R = aX + bY

(Equation 3)
E (R) = E (aX + bY) = E (aX) + E (bY)
= AE (X) + bE (Y)

(Equation 4)
Var (R) = E (R−E (R)) ²
= AE (X-E (X)) ² + aE (Y-E (Y)) ²
+ 2abE (X-E (X)) E (Y-E (Y))

  Here, the expected return E (R) of Equation 3 is given by multiplying the expected returns E (X) and E (Y) of each case by the respective investment distributions a and b. Also, the variance Var (R) of Equation 4 is twice the variances of the cases A and B, which is obtained by multiplying the variances Var (X) and Var (Y) of each case by the respective investment allocations a and b. It is given by adding the product of the respective investment allocations a and b. Here, when the cases are independent from each other, the covariance is 0, and therefore, Equation 4 is simplified to Equation 5.

(Equation 5)
Var (R) = aE (X-E (X)) ² + aE (Y-E (Y)) ²

  Therefore, even if the number of cases is three or more, assuming that the profits of each case are independent, the number of cases is applied to three or more as it is according to Equations 3 and 5.

  In addition, the 95% VaR value of the probability distribution in the entire portfolio is obtained by summing up the 95% VaR values calculated for each individual case.

  Next, the portfolio analysis device 2 reproduces the probability density distribution or probability distribution (cumulative value) shape of the entire portfolio from the probability density distribution shape for each case reproduced by the data reproduction unit 16 as a graph, and displays the computer 5 display. It is displayed on the screen of the device (# 13).

  Next, the portfolio analysis apparatus 2 accepts input of the set value of the risk upper limit value (management frame) by the user's input operation, and determines whether the risk of the entire portfolio exceeds the management frame in each fiscal year. The result is displayed on the screen of the display device of the computer 5 (# 14).

Second Embodiment
Next, a second embodiment of the device of the present invention will be described. As shown in FIG. 6, the present invention device 6 of the second embodiment is configured on a computer 5 as a part of the present invention system 4 together with the portfolio analysis device 2 and the simulation device 3 as in the first embodiment. The software processing apparatus manages risk evaluation index value data, which is a set of risk evaluation index values (cash flows) generated by Monte Carlo simulation, on the computer 5.

  Further, as shown in FIG. 6, the device 6 of the present invention includes a data input unit 21, a set value storage unit 22, a set value input unit 23, a risk evaluation index value database 24, a probability density calculation unit 25, A data storage unit 26 and a data reproduction unit 27 are provided. Here, the data input unit 21, the set value input unit 23, the probability density calculation unit 25, the data storage unit 26, and the data reproduction unit 27 execute a program for processing each unit separately on the computer 5. It is realized by. Further, the set value storage unit 22 and the risk evaluation index value database 24 are respectively constructed in a predetermined storage area of a predetermined storage device provided in the computer 5. The set value storage unit 22 and the risk evaluation index value database 24 may be configured in separate storage devices.

  The difference from the first embodiment is that a probability density calculation unit 25 is provided. Due to this difference, the data structure of the risk evaluation index value database 24 and the details of the data storage unit 26 and the data reproduction unit 27 are processed. The contents are different from the first embodiment. Specifically, step # 7 in the flowchart shown in FIG. 2 is different from step # 11 in the flowchart shown in FIG.

  Hereinafter, operations of the probability density calculation unit 25 and the data storage unit 26 in step # 7 will be described.

  First, the probability density calculation unit 25 extracts only the percentile value (risk evaluation index value) in each of the above percentile setting values from the risk evaluation index value data generated in step # 3, and the data of the first embodiment. The probability density is calculated in exactly the same manner as the reproduction unit 16 calculates the probability density at each percentile setting value based on Equation (1). The data storage unit 26 uses the probability density calculated by the probability density calculation unit 25 and the minimum and maximum values of the risk evaluation index value as the risk evaluation index instead of the percentile value (risk evaluation index value) in each percentile setting value. Save in the value database 14. Similarly, for the NPV (net present value) generated in step # 4, the probability density at the same percentile setting value is calculated and stored. In addition, the average value and variance of the risk evaluation index value data, and the 95% VaR value for each year are stored in the same manner as in the first embodiment. Therefore, in the second embodiment, the risk evaluation index value database 14 stores the probability density at each percentile setting value calculated in advance.

  Next, in step # 11 in the business risk analysis processing for the entire portfolio shown in the flowchart shown in FIG. 3, the risk evaluation index value data extracted in step # 10 includes the already calculated probability density. Since the reproduction unit 27 does not need to calculate the probability density based on Equation 1, in the second embodiment, in step # 11, the probability density and the minimum value in each percentile setting value included in the extracted risk evaluation index value data. The probability density distribution may be reproduced using the value and the maximum value.

  Regarding the other processes, the first embodiment and the second embodiment are the same, and thus redundant description is omitted.

  Hereinafter, another embodiment will be described.

  <1> In each of the above embodiments, in setting the percentile set value in step # 6 of the risk and return analysis process for each individual case shown in the flowchart of FIG. Although the description has been given of the mode in which the data storage unit 15 receives the input setting value and reads out and uses the data stored in the setting value storage unit 12, the percentile setting value does not use the setting input value from the user. It is also preferable to automatically set the percentile setting value so as to obtain an appropriate thinning interval in accordance with the distribution shape of the risk evaluation index value data generated in step # 3. In this case, the device 1 of the present invention includes a percentile setting unit that automatically sets the percentile setting value instead of or in addition to the setting value input units 13 and 23, and the percentile setting unit distributes the risk evaluation index value data. The shape is analyzed, and it is preferable that the setting is made so that the interval between the percentile setting values becomes fine in a range where the probability density is less than or equal to a predetermined value and a range where the probability density is changing rapidly. In this way, it is possible to perform reproducibility with high accuracy and a large amount of data reduction with respect to risk evaluation index value data having an arbitrary distribution shape.

  <2> In each of the embodiments described above, the embodiment in which the device 1 of the present invention is integrated with the portfolio analysis device 2 and the simulation device 3 to configure the system 4 of the present invention has been described. You may isolate | separate from the analyzer 2 and the simulation apparatus 3, and may be provided independently. In addition, the data input unit 11, the set value storage unit 12, the set value input unit 13, the risk evaluation index value database 14, and the data storage unit 15 of the device 1 of the present invention are integrally configured, and the data reproduction unit 16 is separated. You may make it comprise with a body. Similarly, the data input unit 21, the set value storage unit 22, the set value input unit 23, the risk evaluation index value database 24, the probability density calculation unit 25, and the data storage unit 26 of the device 6 of the present invention are integrally configured. The data reproduction unit 27 may be configured separately. In this case, the data reproduction units 16 and 27 may be provided on the portfolio analyzer 2 side.

  <3> The processing content of the portfolio analysis apparatus 2 in each of the above embodiments is an example, and is not limited to the processing content of the above embodiment.

The block diagram which shows schematic structure of 1st Embodiment of the risk evaluation index value data management apparatus and risk management system which concern on this invention. The flowchart which shows the operation | movement procedure of the risk and return analysis process for every individual case in one Embodiment of the risk evaluation index value data management apparatus and risk management system which concern on this invention The flowchart which shows the operation | movement procedure of the business risk analysis process in the whole portfolio in one Embodiment of the risk evaluation index value data management apparatus and risk management system which concern on this invention Diagram showing the data structure of the risk assessment index value database Diagram explaining the probability density distribution of risk assessment index values and the algorithm for calculating the probability density and the probability density calculation algorithm The block diagram which shows schematic structure of 2nd Embodiment of the risk evaluation index value data management apparatus and risk management system which concern on this invention.

Explanation of symbols

1, 6: Risk evaluation index value data management device according to the present invention 2: Portfolio analysis device 3: Simulation device 4: Risk management system according to the present invention 5: Computer 11, 21: Data input unit 12, 22: Setting value storage Units 13, 23: set value input units 14, 24: risk evaluation index value database 15, 26: data storage unit 16, 27: data reproduction unit 25: probability density calculation unit

Claims (9)

A risk evaluation index value data management device for managing risk evaluation index value data, which is a set of discrete index values generated by Monte Carlo simulation for evaluating business risk, on a computer,
A data input unit that accepts input of the risk assessment index value data and temporarily stores it in a predetermined storage area;
A set value storage unit that stores discrete percentile setting values of the occurrence probability of the risk evaluation index value data input to the data input unit;
A data storage unit that extracts the percentile value, the maximum value, and the minimum value in the percentile setting value of the risk evaluation index value from the temporarily stored risk evaluation index value data and stores it in a risk evaluation index value database;
A risk evaluation index value data management device comprising:   The percentile value, the maximum value, and the minimum value of the risk evaluation index value are extracted from the risk evaluation index value database, and each percentile setting is based on the extracted percentile value, the maximum value, and the minimum value. The risk evaluation index value data management apparatus according to claim 1, further comprising a data reproduction unit that calculates a probability density in the value and reproduces the probability density distribution of the risk evaluation index value. A risk evaluation index value data management device for managing risk evaluation index value data, which is a set of discrete index values generated by Monte Carlo simulation for evaluating business risk, on a computer,
A data input unit that accepts input of the risk assessment index value data and temporarily stores it in a predetermined storage area;
A set value storage unit that stores discrete percentile setting values of the occurrence probability of the risk evaluation index value data input to the data input unit;
The percentile value, maximum value, and minimum value in the percentile setting value of the risk evaluation index value are extracted from the temporarily stored risk evaluation index value data, and based on the extracted percentile value, the maximum value, and the minimum value A probability density calculator for calculating a probability density at each percentile setting value,
A data storage unit that stores the probability density, the maximum value, and the minimum value in the percentile setting value of the risk evaluation index value in a risk evaluation index value database;
A risk evaluation index value data management device comprising:   The probability density, the maximum value, and the minimum value in the percentile setting value of the risk evaluation index value are extracted from the risk evaluation index value database, and the extracted probability density, the maximum value, and the minimum value are extracted. 4. The risk evaluation index value data management apparatus according to claim 3, further comprising a data reproduction unit that reproduces a probability density distribution of the risk evaluation index value based on the data.   The risk evaluation index value according to any one of claims 1 to 4, wherein the percentile setting value is set so as to be subdivided so as to increase toward the bottom of a probability density distribution having a smaller probability density. Data management device.   The risk evaluation index value data management apparatus according to claim 1, further comprising a setting value input unit that receives an input of the percentile setting value and stores the setting value in a setting value storage unit. . The risk evaluation index value data management device according to any one of claims 1 to 6,
A portfolio analysis device that performs risk and return analysis for each project in the portfolio and business risk analysis for the entire portfolio by software processing of a computer using a predetermined analysis index;
A simulation device for generating the risk evaluation index value by Monte Carlo simulation based on a plurality of scenarios;
A risk management system comprising:   The risk evaluation index value data management which has a computer program which performs each process of the said data input part in the risk evaluation index value data management apparatus of Claim 2 in the said data storage part and the said data reproduction part on a computer, respectively program.   5. A computer program for executing each processing of the data input unit, the probability density calculation unit, the data storage unit, and the data reproduction unit in the risk evaluation index value data management device according to claim 4 on a computer. Risk assessment index value data management program.

Images