JP5911438B2 - Method selection method and program, and information processing apparatus - Google Patents

Description

  The present technology relates to a technology for estimating a curve representing a relationship between an interest rate and a remaining period.

  A yield curve is a curve representing the relationship between the remaining period of a discount bond (remaining time until maturity date) and the average interest rate. In the case of Japan, there is no discount bond for government bonds, but there is a method for estimating the yield curve using the transaction price of interest-bearing government bonds (ie market price) instead of discount bonds. In countries where government bonds, which are discount bonds, exist, there are many methods for estimating the yield curve using the transaction price of the government bonds.

  An appropriate yield curve cannot be obtained unless an appropriate method is used according to market conditions and bond differences, but until now it has been difficult to select an appropriate method from among several methods. There are approximately 300 Japanese government bonds, with a maximum maturity of 40 years. However, there is a problem that the number of issues of government bonds (for example, 30 to 40 years) with a long term to maturity is small. This is because the maximum period until maturity was 30 years until October 2007, but a government bond with a maturity of 40 years was issued in November 2007, and the number of issues was between 30 and 40 years. There are few blank periods. Since this blank period exists, it is difficult to estimate the yield curve.

Kawasaki Noritori, Yasuchi Tomohiro, Yield Curve Estimation by Regularized Nonlinear Regression Model, Statistical Mathematics, Vol. 50, No. 2, 149-164, 2002 Mizuho Report “Understanding the Structure of Government Bond Spot Rate Curves by Principal Component Analysis and Examining its Predictability: Using a Common Factor Model Based on Macroeconomic and Financial Variables” published on September 21, 2010, Mizuho Research Institute James M, Steeley, "Estimating the gilt-edged term structure basis splines and confidence intervals", Journal of Business Finance & Accounting, 18 (4), June 1991 Takuya Abe, Masanobu Sasaki, Hidetoshi Matsuoka, Hiroshi Ikeda, “Estimation of yield curves using nonparametric methods and inter-market data”, Japan Finance, Securities Metrology and Engineering Society (JAFEE), August 3, 2012

  Therefore, the objective of this technique is to provide the technique for enabling selection of the method which can estimate an appropriate yield curve from the transaction price of a bond according to one side.

  The method selection method according to the present technology is as follows: (A) In a data storage unit that stores bond data including the maturity date and price of the bond for each bond, one of the bonds whose remaining period falls within a predetermined period of the remaining period. (B) A curve representing the relationship between the interest rate and the remaining period according to each of a plurality of predetermined methods using bond data of at least some bonds among bonds other than the selected bonds. And (C) calculating a price of the selected bond from the estimated curve and bond data of the selected bond for each of the plurality of predetermined methods, and (D) for each of the plurality of predetermined methods. Calculating an evaluation value based on the difference between the price included in the bond data of the selected bond and the calculated price, and (E) based on the evaluation value of each of the plurality of predetermined methods. Select one of the predetermined methods It includes a process for.

  A method that can estimate an appropriate yield curve from the transaction price of a bond can be selected.

FIG. 1 is a diagram illustrating a configuration example of an information processing apparatus according to an embodiment. FIG. 2 is a diagram illustrating an example of data stored in the bond DB. FIG. 3 is a diagram showing a main processing flow according to the embodiment. FIG. 4 is a diagram for explaining the data selection process. FIG. 5 is a diagram for explaining the data selection process. FIG. 6 is a diagram for explaining the data selection process. FIG. 7 is a diagram for explaining the data selection process. FIG. 8 is a diagram for explaining the data selection process. FIG. 9 is a diagram for explaining the data selection process. FIG. 10 is a diagram for explaining the data selection process. FIG. 11 is a diagram for explaining the data selection process. FIG. 12 is a diagram for explaining the data selection process. FIG. 13 is a diagram showing a processing flow of data selection processing. FIG. 14 is a diagram illustrating an example of an estimated yield curve. FIG. 15 is a diagram illustrating an example of the estimated price. FIG. 16 is a diagram for explaining the price difference. FIG. 17 is a diagram illustrating a main processing flow according to the embodiment. FIG. 18 is a diagram illustrating an example of data stored in the third data storage unit. FIG. 19 is a functional block diagram of a computer.

  FIG. 1 shows a configuration example of an information processing apparatus 100 according to the present embodiment. The information processing apparatus 100 includes a bond database (DB) 101, a data extraction unit 102, a first data storage unit 103, a control unit 104, a second data storage unit 105, an evaluation index calculation unit 106, and a third Data storage unit 107, selection unit 108, setting data storage unit 109, input unit 112, fourth data storage unit 110, output unit 111, first yield curve calculation unit 120a, second yield curve calculation unit 120b,. . . An n-th yield curve calculation unit 120n.

  The bond DB 101 stores basic data of each bond (maturity date, face value, interest payment amount (or interest rate), interest payment date data, etc.) and transaction price data of each bond. The transaction price data includes data of the transaction history of each bond (here, the transaction price for each transaction day, but may be the transaction price for each other unit such as each time). For example, data is added to the transaction price data for each transaction date. An example of data stored in the bond DB 101 is shown in FIG. In the example of FIG. 2, basic data including maturity date, interest payment rate, face value, and payment interval, and transaction price data including market price, transaction date, and transaction volume are registered for each issue. Yes.

  The setting data storage unit 109 stores data of a blank period T ([t1, t2]) input via the input unit 112. The data extraction unit 102 extracts bond data used for selecting the yield curve calculation method from the bond DB 101 according to the data stored in the setting data storage unit 109 and stores the bond data in the first data storage unit 103.

  The control unit 104 controls the first yield curve calculation unit 120a to the nth yield curve calculation unit 120n. More specifically, the control unit 104 generates, for example, a yield curve for each trading day from the bond data stored in the first data storage unit 103 in the first yield curve calculation unit 120a to the nth yield curve calculation unit 120n. The yield curve data is stored in the second data storage unit 105. When an appropriate yield curve calculation method is selected, the control unit 104 reads bond data from the bond DB 101, and sends the bond data from the read bond data to the yield curve calculation unit 120 for the appropriate yield curve calculation method. The yield curve is calculated, and the yield curve data is stored in the fourth data storage unit 110.

  The evaluation index calculation unit 106 uses the bond data stored in the first data storage unit 103 and the yield curve data stored in the second data storage unit 105 to estimate transactions for each transaction day for each yield curve. Calculate the price. Then, the evaluation index calculation unit 106 calculates an evaluation index value based on the difference between the estimated transaction price for each transaction day and the actual transaction price for each yield curve calculation method, and stores the evaluation index value in the third data storage unit 107.

  The selection unit 108 selects the most appropriate yield curve calculation method from the evaluation values of the respective yield curve calculation methods stored in the third data storage unit 107.

  The output unit 111 outputs appropriate yield curve data stored in the fourth data storage unit 110 to an output device (for example, a display device, a printing device, or another computer connected to a network).

  The first yield curve calculation unit 120a to the nth yield curve calculation unit 120n are processing units that calculate yield curves using different yield curve calculation methods. For example, the methods shown in Non-Patent Documents 1 to 4 are used.

  Next, processing contents of the information processing apparatus 100 will be described with reference to FIGS. 3 to 18. The data extraction unit 102 of the information processing apparatus 100 detects the determination timing of the yield curve calculation method from the data content of the bond DB 101 or the input from the input unit 112 or the like (FIG. 3: step S1).

  For example, the timing has a remaining period that falls into the blank period T of the remaining period when a bond stock having a remaining period that falls into the blank period T (= (t1, t2]) of the remaining period is added. This may be the case when a predetermined period has elapsed since the bond issue was added (for example, after the number of transaction days (more specifically, the transaction price) has increased sufficiently), or when instructed by the user.

  When such determination timing is detected, the data extraction unit 102 executes a data selection process for extracting bond data from the bond DB 101 (step S3). This data selection process will be described with reference to FIGS.

  For example, as shown in FIG. 4, when a first 40-year bond is issued (that is, trading date X1), a bond brand having a remaining period that falls into the blank period T is added. In the following, the date on which the second 40-year bond was issued is trading day X2, the date on which the third 40-year bond was issued is trading date X3, and the fourth 40-year bond is It is assumed that the date of issue is the transaction date X4. In the following description, it is assumed that t1 = 30 years and t2 = 40 years. In FIG. 4, it is assumed that time flows from left to right. That is, on the transaction date X1, the transaction price is determined for the bond brand with a remaining period of 40 years. In such a case, bond data D1 is extracted for each transaction date prior to the transaction date X1. In this case, the setting data storage unit 109 defines in advance how far back the transaction date is.

Furthermore, in the present embodiment, the number nT of bonds having a remaining period that falls within the blank period _T is counted on the previous trading day of the trading day X1. Here, n _T = 0. And in the bond data D1 of each trading day prior to the trading date X1 (FIG. 5. One graph in FIG. 5 represents data for one trading day, and x and circles represent brands). Bonds having a remaining period that falls within the period T ₀ (= (t0 (= t1- (t2-t1)), t1]). In this example, (20, 30]). Of these, stocks other than n _T +1 (in this case, 1) stocks with the longer remaining period (circled in Fig. 5) are excluded from stocks used in determining the yield curve calculation method. Excluded bond issues are subject to comparison between the calculated price and the actual transaction price.

For example, as shown in FIG. 6, when a second 40-year bond is issued, a bond brand having a remaining period that falls into the blank period T is added. Also in FIG. 6, time flows from left to right. That is, on the transaction date X2, the transaction price is determined for a bond brand having a remaining period of 40 years and a bond brand having a remaining period slightly shorter than 40 years. In such a case, n _T = 1, and the bond data D2 of each transaction date after the transaction date X1 before the transaction date X2 is extracted. That is, extracts the bonds data D2 for each trading day the number n _T of claims stocks with remaining period as fall blank period T is not changed.

Then, bond data D2 for each trading day after trading date X1 before trading date X2 (FIG. 7. One graph in FIG. 7 represents data for one trading day, and triangles, x and circles indicate stocks. Among the issues of bonds that have a remaining period that falls within the period T ₀ in terms of remaining period, other than issues of n _T +1 (in this case, 2) bonds from the longer remaining period The issues (circles in FIG. 7) are excluded from the issues of bonds used to determine the yield curve calculation method. Excluded bond issues are subject to comparison between the calculated price and the actual transaction price. In addition, a bond brand (triangle mark in FIG. 7) having a remaining period that falls within the blank period T is completely excluded from the processing target.

As described above, when there are not enough bonds (for example, three) for the blank period T as they are at the time of processing, a blank period is created in a pseudo manner. A blank period is set in the period T ₀ having the same length.

Further, for example, as shown in FIG. 8, when a third 40-year bond is issued, a bond issue having a remaining period that falls into the blank period T is added. Also in FIG. 8, time flows from left to right. That is, on the transaction date X3, the transaction price is determined for two issues: a bond issue with a remaining period of 40 years and a bond with a slightly shorter remaining period of 40 years. In such a case, n _T = 2 and the bond data D3 of each transaction date after the transaction date X2 before the transaction date X3 is extracted. That is, extracts the bonds data D3 for each trading day the number n _T of bond issues with remaining period as fall blank period T is not changed.

When n _T becomes 2, bond data D3 for each trading day before trading date X3 and after trading date X2 (FIG. 9; one graph in FIG. 9 represents data for one trading day, x and Among the issues of bonds that have a remaining period that falls within the period T for the remaining period, the one with the shorter remaining period (circled in FIG. 9) Excluded from bonds that are used to determine the yield curve calculation method. Excluded bond issues are subject to comparison between the calculated price and the actual transaction price.

Furthermore, for example, as shown in FIG. 10, when a fourth 40-year bond is issued, a brand of a bond having a remaining period that falls into the blank period T is added. Also in FIG. 10, time flows from left to right. In other words, on the transaction date X4, the transaction price is determined for three issues: a bond issue with a remaining period of 40 years and a bond with a slightly shorter remaining period of 40 years. In such a case, n _T = 3, and the bond data D4 for each transaction date after the transaction date X3 before the transaction date X4 is extracted. That is, extracts the bonds data D4 for each trading day the number n _T of bond issues with remaining period as fall blank period T is not changed.

Furthermore, since n _T is 2 or more, bond data D4 for each trading day before trading date X4 and after trading date X3 (FIG. 11. One graph in FIG. 11 shows data for one trading day. Among the issues of bonds having a remaining period that falls within the period T of the remaining period, one symbol (indicated by a circle in FIG. 11) from the shorter remaining term. ) Is excluded from bond issues used to determine the yield curve calculation method. Excluded bond issues are subject to comparison between the calculated price and the actual transaction price.

  Not only the timing of issuing 40-year bonds as described above, but also, for example, if bond data D3 or D4 contains a sufficient number of transaction date data, not only the timing of bond issuance. The same data extraction processing may be used at other timings.

  For example, as shown in FIG. 12, there are a sufficient number of transaction days after transaction date X3 when the third 40-year bond was issued at the time of processing prior to the time when the fourth 40-year bond was issued. If it is equal to or greater than (for example, the number specified in the setting data storage unit 109), bond data D5 on the transaction date after the transaction date X3 and before the processing time may be extracted. The subsequent processing may be the same as the processing for D4.

The contents of such processing will be specifically described with reference to FIG. The data extraction unit 102 counts the number of issues n _T in the period T stored in the setting data storage unit 109 in the data of the previous transaction date at the time of processing, and the bond on the transaction date in a range where the number of issues n _T does not change. Data is extracted from the bond DB 101 and stored in the first data storage unit 103 (FIG. 13: step S51).

Then, the data extraction unit 102 determines whether n _T is greater than 1 (step S55). If n _T is 1 or less, as shown in FIG. 4 to FIG. 7, the data extraction unit 102 determines whether the period T ₋₁ (period before t0 = [0, t0]) includes the remaining period. , N _T +1 brands having the longer remaining period among the brands including the remaining period in T ₀ are set as the yield curve estimation target Vp (step S57).

Further, the data extraction unit 102 sets, among the brands included in the period T ₀ , brands that are not included in the yield curve estimation target Vp (excluding brands included in the period T) as the price calculation target Vc (step S1). S59). Then, the process returns to the caller process.

On the other hand, when n _T exceeds 1, the data extraction unit 102, as shown in FIGS. 8 to 12, the brand including the remaining period in the periods T ₋₁ and T ₀ and the remaining period in the period T. Of the included brands, brands other than the one with the shortest remaining period are set as the yield curve estimation target Vp (step S61).

  Further, the data extraction unit 102 sets a brand that is not included in the yield curve estimation target Vp among the brands included in the remaining period in the period T as the price calculation target Vc (step S63). Then, the process returns to the caller process.

By executing such processing, the bond brands can be divided into yield curve estimation targets Vp and price calculation targets Vc in the blank period T or a period T ₀ corresponding to a pseudo blank period. It becomes like this.

Note that the yield curve estimation target Vp has a small number of issues in the blank period T or the period T ₀ so that it can be determined whether or not an appropriate yield curve can be estimated even in such a state.

  Returning to the description of FIG. 3, the control unit 104 identifies one unevaluated method among the n yield curve calculation methods (step S <b> 5). And the control part 104 extracts the bond data of an unprocessed transaction day from the bond data of the brand selected as the yield curve estimation target Vp in the data selection process from the first data storage part 103 (step S7). . Then, the control unit 104 outputs a request for yield curve estimation including the extracted bond data to the yield curve calculation unit 120 according to the method specified in step S5, and the yield curve calculation unit 120 outputs the request to the yield curve calculation unit 120. Calculation processing is performed (step S9). Upon receiving the yield curve data as the processing result from the requested yield curve calculation unit 120, the control unit 104 stores the data in the second data storage unit 105.

  Thereafter, the evaluation index calculation unit 106 uses the yield curve data stored in the second data storage unit 105 and the bond data (price calculation target Vc) stored in the first data storage unit 103, The estimated price of each brand of the price calculation target Vc is calculated (step S11).

For bond data, basic data is mainly used. r (t) is the yield curve, t _{1 to} t _L is the time when the cash flow of the issue occurs, L is the number of future cash flows, R is the redemption amount of the issue, c is the amount of interest paid once, If Pc is an estimated price, the estimated price is expressed as follows.

  When a yield curve as schematically shown in FIG. 14 is obtained, an estimated price (square mark) as schematically shown in FIG. 15 is calculated.

Then, the evaluation index calculation unit 106 calculates the square value of the difference between the actual price of each brand of the price calculation target Vc and the estimated price corresponding to each brand (step S13). As schematically shown in FIG. 16, the square value of the difference e _ij between the estimated price of the square mark and the actual price represented by the x mark is calculated. Note that e _ij represents the price difference of the j-th issue on the i-th transaction date. Then, the processing shifts to the processing in FIG.

  17, the control unit 104 determines whether there is an unprocessed transaction date among the bond data stored in the first data storage unit 103 (step S15). If there is an unprocessed transaction date, the process returns to step S7 in FIG.

  On the other hand, if there is no unprocessed transaction date, the evaluation index calculation unit 106 calculates an evaluation value for the yield curve calculation method specified in step S5 based on the average of the square values of the price differences, and the third The data is stored in the data storage unit 107 (step S17). For example, data as illustrated in FIG. 18 is stored in the third data storage unit 107. In the example of FIG. 18, an evaluation value is registered for each yield curve calculation method.

  The average value of the squares of the price differences is an example of the evaluation value, and may be the sum of the squares of the price differences, the average value of the absolute values of the price differences, It may be the sum of absolute values. Furthermore, the reciprocal of these values may be adopted as the evaluation value to increase the value.

  Then, the control unit 104 determines whether there is an unprocessed yield curve calculation method (step S19). If there is an unprocessed yield curve calculation method, the process returns to step S5 in FIG.

  On the other hand, when there is no unprocessed yield curve calculation method, the selection unit 108 selects the yield curve calculation method with the smallest evaluation value as an appropriate yield curve calculation method (step S21), and the control unit 104 Notifies the identifier of the selected yield curve calculation method. In the example of FIG. 18, since the evaluation value of method 1 is the minimum, method 1 is selected.

  Then, the control unit 104 extracts bond data at the time of processing from the bond DB 101, and outputs the bond data to the yield curve calculation unit 120 according to the selected yield curve calculation method so that the yield curve is calculated (step S23). ). Upon receiving the yield curve data from the yield curve calculation unit 120, the control unit 104 stores the data in the fourth data storage unit 110.

  The output unit 111 outputs the yield curve data stored in the fourth data storage unit 110 to the output device (step S25).

  By performing such processing, an appropriate yield curve calculation method can be selected at any given time.

Although the embodiment of the present technology has been described above, the present technology is not limited to this. For example, the functional block diagram of FIG. 1 is an example, and may not match the program module configuration. As for the processing flow, as long as the processing result does not change, the processing order may be changed or a plurality of processing steps may be executed in parallel. In particular, since the first yield curve calculation unit 120a to the nth yield curve calculation unit 120n can be processed separately, the calculation of the yield curve may be performed in parallel. Furthermore, _T is based on n _T on the previous transaction date at the time of processing (however, when the transaction is performed every day. Generally, there is no price of additional issued bonds such as one unit time ago). Although an example using any of T ₀ has been shown, there is a case where n _T at the time of processing is used as a reference. As described above, since n _T does not change to some extent, n _T may be counted on a transaction date determined based on the processing time, such as several days before the processing time. It should be noted that the case where n _T is counted at the time of processing can be handled by changing the threshold value. Moreover, the threshold value described above is an example, and another threshold value may be adopted. Further, when the process of selecting a method at the timing designated by the user is executed, a setting may be made such that the process is not performed if n _T = 0.

  The information processing apparatus 100 described above is a computer apparatus, and as shown in FIG. 19, a memory 2501, a CPU (Central Processing Unit) 2503, a hard disk drive (HDD: Hard Disk Drive) 2505, and a display device. A display control unit 2507 connected to 2509, a drive device 2513 for the removable disk 2511, an input device 2515, and a communication control unit 2517 for connecting to a network are connected by a bus 2519. An operating system (OS) and an application program for executing the processing in this embodiment are stored in the HDD 2505, and are read from the HDD 2505 to the memory 2501 when executed by the CPU 2503. The CPU 2503 controls the display control unit 2507, the communication control unit 2517, and the drive device 2513 according to the processing content of the application program, and performs a predetermined operation. Further, data in the middle of processing is mainly stored in the memory 2501, but may be stored in the HDD 2505. In an embodiment of the present technology, an application program for performing the above-described processing is stored in a computer-readable removable disk 2511 and distributed, and installed from the drive device 2513 to the HDD 2505. In some cases, the HDD 2505 may be installed via a network such as the Internet and the communication control unit 2517. Such a computer apparatus realizes various functions as described above by organically cooperating hardware such as the CPU 2503 and the memory 2501 described above and programs such as the OS and application programs. .

  The above-described embodiment can be summarized as follows.

  The method selection method according to the present technology is as follows: (A) In a data storage unit that stores bond data including the maturity date and price of the bond for each bond, one of the bonds whose remaining period falls within a predetermined period of the remaining period. (B) A curve representing the relationship between the interest rate and the remaining period according to each of a plurality of predetermined methods using bond data of at least some bonds among bonds other than the selected bonds. And (C) calculating a price of the selected bond from the estimated curve and bond data of the selected bond for each of the plurality of predetermined methods, and (D) for each of the plurality of predetermined methods. Calculating an evaluation value based on the difference between the price included in the bond data of the selected bond and the calculated price, and (E) based on the evaluation value of each of the plurality of predetermined methods. Select one of the predetermined methods It includes a process for.

  In this way, for example, an appropriate predetermined method for estimating a curve representing the relationship between the interest rate and the remaining period can be selected in consideration of the predetermined period with a small number of applicable bonds.

  The processing for selecting a bond described above is (a1) depending on the number of bonds whose remaining period falls within the first predetermined period of the remaining period at the processing point or transaction point determined based on the processing point in the data storage unit. A portion of bonds that have a remaining period in the first predetermined period, and a part of bonds that have a remaining period in the second predetermined period that has the same length as the first period. A process for selecting one of the above may be included. For example, even when a sufficient number of bonds do not exist in the first predetermined period corresponding to the blank period, it is possible to generate a state in which a blank period exists in a pseudo manner by the second predetermined period.

  In addition, the above-described processing for selecting a bond includes (a2) a process of extracting bond data for a transaction within a period in which the number of bonds whose remaining period falls within the first predetermined period before the processing time does not change. It may be included. This is for performing stable curve estimation.

  Furthermore, the price mentioned above may be the price of each trading day. In this case, the process for estimating the curve described above may include a process for estimating a curve representing the relationship between the spot rate and the remaining period for each transaction day. Further, the processing for calculating the price of the bond described above may include processing for calculating the price of the selected bond from the curve of the transaction date and all the selected case data for each transaction date. . Then, the process for calculating the evaluation value described above is performed based on the difference between the selected price and the price included in the bond data of the selected bond for each of the plurality of predetermined methods for each trading day. And a process of calculating a second evaluation value from the first evaluation value for each of a plurality of predetermined methods. In this way, evaluation can be performed with more samples.

  Furthermore, in the above-described processing for selecting a bond, a bond having a short remaining period may be preferentially selected. Bonds for price comparison can be selected appropriately.

  Note that a program for causing a computer to execute the processing described above can be created, and the program includes, for example, a flexible disk, an optical disk such as a CD-ROM, a magneto-optical disk, and a semiconductor memory (for example, ROM). Or a computer-readable storage medium such as a hard disk or a storage device. Note that data being processed is temporarily stored in a storage device such as a RAM.

  The following supplementary notes are further disclosed with respect to the embodiments including the above examples.

(Appendix 1)
In the data storage unit that stores bond data including the maturity date and price of the bond for each bond, select a part of bonds that have a remaining period within a predetermined period of the remaining period,
Estimating a curve representing the relationship between the interest rate and the remaining period according to each of a plurality of predetermined methods using bond data of at least some of the bonds other than the selected bonds,
For each of the plurality of predetermined methods, a price of the selected bond is calculated from the estimated curve and bond data of the selected bond.
For each of the plurality of predetermined methods, calculate an evaluation value based on a difference between the price included in the bond data of the selected bond and the calculated price,
A method selection method executed by a computer, including a process of selecting any one of the plurality of predetermined methods based on an evaluation value of each of the plurality of predetermined methods.

(Appendix 2)
The process of selecting the bond is:
In the data storage unit, the remaining period enters the first predetermined period according to the number of bonds whose remaining period falls in the first predetermined period for the remaining period at the processing time or the transaction time determined based on the processing time. Supplementary note 1 including a process of selecting one of the bonds and one of the bonds having a remaining period in the second predetermined period having the same length as the first predetermined period The method selection method described.

(Appendix 3)
The process of selecting the bond is:
The method selection method according to supplementary note 2, including a process of extracting bond data for a transaction within a period in which the number of bonds whose remaining period falls within the first predetermined period before the processing time point does not change.

(Appendix 4)
The price is the price of each trading day,
The process of estimating the curve includes a process of estimating a curve representing the relationship between the spot rate and the remaining period for each transaction day,
The process of calculating the price of the bond includes a process of calculating the price of the selected bond from the curve of the transaction date and all selected data for each transaction day,
The process of calculating the evaluation value includes
For each of the plurality of predetermined methods, for each trading day, a first evaluation value based on a difference between the selected price and a price included in bond data of the selected bond is calculated,
The method selection method according to any one of supplementary notes 1 to 3, further comprising: calculating a second evaluation value from the first evaluation value for each of the plurality of predetermined methods.

(Appendix 5)
The method selection method according to claim 2, wherein, in the process of selecting the bond, the bond having a short remaining period is preferentially selected.

(Appendix 6)
In the data storage unit that stores bond data including the maturity date and price of the bond for each bond, select a part of bonds that have a remaining period within a predetermined period of the remaining period,
Estimating a curve representing the relationship between the interest rate and the remaining period according to each of a plurality of predetermined methods using bond data of at least some of the bonds other than the selected bonds,
For each of the plurality of predetermined methods, a price of the selected bond is calculated from the estimated curve and bond data of the selected bond.
For each of the plurality of predetermined methods, calculate an evaluation value based on a difference between the price included in the bond data of the selected bond and the calculated price,
A method selection program for causing a computer to execute a process of selecting any one of the plurality of predetermined methods based on an evaluation value of each of the plurality of predetermined methods.

(Appendix 7)
A first selection unit that selects a portion of bonds out of bonds whose remaining period falls within a predetermined period of the remaining period in a data storage unit that stores bond data including a maturity date and a price of the bond for each bond;
An estimation unit that estimates a curve representing a relationship between an interest rate and a remaining period according to each of a plurality of predetermined methods, using bond data of at least some bonds among bonds other than the selected bonds;
The price of the selected bond is calculated from the estimated curve and the bond data of the selected bond for each of the plurality of predetermined methods, and the selected for each of the plurality of predetermined methods. A calculation unit that calculates an evaluation value based on a difference between the price included in the bond data of the bond and the calculated price;
A second selection unit that selects any one of the plurality of predetermined methods based on an evaluation value of each of the plurality of predetermined methods;
An information processing apparatus including:

101 Bond DB
102 data extraction unit 103 first data storage unit 104 control unit 105 second data storage unit 106 evaluation index calculation unit 107 third data storage unit 108 selection unit 109 setting data storage unit 110 fourth data storage unit 111 output unit 112 input unit 120 Yield curve calculator

Claims (7)

In the data storage unit that stores bond data including the maturity date and price of the bond for each bond, select a part of bonds that have a remaining period within a predetermined period of the remaining period,
Estimating a curve representing the relationship between the interest rate and the remaining period according to each of a plurality of predetermined methods using bond data of at least some of the bonds other than the selected bonds,
For each of the plurality of predetermined methods, a price of the selected bond is calculated from the estimated curve and bond data of the selected bond.
For each of the plurality of predetermined methods, calculate an evaluation value based on a difference between the price included in the bond data of the selected bond and the calculated price,
A method selection method executed by a computer, including a process of selecting any one of the plurality of predetermined methods based on an evaluation value of each of the plurality of predetermined methods. The process of selecting the bond is:
In the data storage unit, the remaining period enters the first predetermined period according to the number of bonds whose remaining period falls in the first predetermined period for the remaining period at the processing time or the transaction time determined based on the processing time. A process of selecting either a part of bonds or a part of bonds having a remaining period in a second predetermined period having the same length as the first predetermined period. The method selection method according to 1. The process of selecting the bond is:
The method selection method according to claim 2, further comprising: extracting bond data for a transaction within a period in which the number of bonds whose remaining period falls within the first predetermined period before the processing time point does not change. The price is the price of each trading day,
The process of estimating the curve includes a process of estimating a curve representing the relationship between the spot rate and the remaining period for each transaction day,
The process of calculating the price of the bond includes a process of calculating the price of the selected bond from the curve of the transaction date and all selected data for each transaction day,
The process of calculating the evaluation value includes
For each of the plurality of predetermined methods, for each trading day, a first evaluation value based on a difference between the selected price and a price included in bond data of the selected bond is calculated,
The method selection method according to any one of claims 1 to 3, further comprising: calculating a second evaluation value from the first evaluation value for each of the plurality of predetermined methods. The method selection method according to claim 2, wherein in the process of selecting a bond, a bond having a short remaining period is selected with priority. In the data storage unit that stores bond data including the maturity date and price of the bond for each bond, select a part of bonds that have a remaining period within a predetermined period of the remaining period,
Estimating a curve representing the relationship between the interest rate and the remaining period according to each of a plurality of predetermined methods using bond data of at least some of the bonds other than the selected bonds,
For each of the plurality of predetermined methods, a price of the selected bond is calculated from the estimated curve and bond data of the selected bond.
For each of the plurality of predetermined methods, calculate an evaluation value based on a difference between the price included in the bond data of the selected bond and the calculated price,
A method selection program for causing a computer to execute a process of selecting any one of the plurality of predetermined methods based on an evaluation value of each of the plurality of predetermined methods. A first selection unit that selects a portion of bonds out of bonds whose remaining period falls within a predetermined period of the remaining period in a data storage unit that stores bond data including a maturity date and a price of the bond for each bond;
An estimation unit that estimates a curve representing a relationship between an interest rate and a remaining period according to each of a plurality of predetermined methods, using bond data of at least some bonds among bonds other than the selected bonds;
The price of the selected bond is calculated from the estimated curve and the bond data of the selected bond for each of the plurality of predetermined methods, and the selected for each of the plurality of predetermined methods. A calculation unit that calculates an evaluation value based on a difference between the price included in the bond data of the bond and the calculated price;
A second selection unit that selects any one of the plurality of predetermined methods based on an evaluation value of each of the plurality of predetermined methods;
An information processing apparatus including:

Images