JP4421276B2 - Execution cost analysis system, execution cost analysis program, and execution cost analysis method - Google Patents

Description

  The present invention relates to an execution cost analysis system, an execution cost analysis program, and an execution cost analysis method for analyzing execution costs in bond trading transactions.

  Conventionally, bonds are bought and sold by relative transactions between investors, including institutional investors such as asset managers, individual investors, etc., and securities companies (dealers), that is, sellers and buyers have a one-to-one relationship. It is done by trading. Therefore, for example, when an investor purchases a bond, by presenting a purchase desired brand to the securities company, a transaction is performed by receiving the price of the desired brand from the securities company. Similarly, when a bond is sold, a transaction is performed in response to the presentation of the price by presenting the securities to be sold to the securities company. Here, even if the brand desired to be purchased and the brand desired to be sold are the same brand, the prices presented by the securities companies are different. That is, generally, a price presented by a securities company includes a brokerage fee (dealer spread) of the securities company for mediating the trading. Therefore, the price at the time of purchase is higher than the price at the time of sale, and the difference between the purchase price and the sale price corresponds to the commission of the securities company.

  In addition, the application of the execution evaluation system which evaluates the transaction transaction (trade execution) of securities is performed by the applicant of this application (refer patent document 1). According to this execution evaluation system, with regard to trading transactions of stocks that are securities, taking into account the unique characteristics of the stocks based on the market conditions at the time of actual trading and the amount of transactions of the trading subject stocks, etc. It is possible to evaluate whether or not sales execution has been properly performed.

JP 2002-140523 A

  By the way, since the buying and selling transactions of bonds are not carried out in a public place like a stock exchange, but are carried out by relative transactions between investors and securities companies, securities companies present them to other investors. It is difficult to know the price. Therefore, there is a problem that it is difficult to obtain materials for determining whether or not the price presented by the securities company is an appropriate price.

  Here, as information on bond trading transactions, the average price of bonds as of 15:00 on the day of the transaction is the Japan Securities Dealers Association (a securities company industry group, and self-regulation) Institution). However, this trading reference statistical value is the average of the median values reported by multiple securities companies, that is, the average value of the median value between the selling price and the purchase price reported by multiple securities companies. is there. Further, the buying / selling reference statistical value is an average value of the middle price only at the time of 15:00 on that day, and it is difficult to obtain information indicating the price movement of the bond from the start of the transaction, that is, the price movement of the bond during the day. As described above, since there is little information to be obtained regarding the buying and selling transaction of bonds, it is difficult to analyze the execution cost in the buying and selling transaction of bonds and evaluate whether or not the selling transaction has been executed properly.

  An object of the present invention is to execute an execution cost analysis system, an execution cost analysis program, and an execution cost analysis that analyze an execution cost of a bond buying and selling transaction and evaluate whether or not a bond buying and selling transaction is appropriately performed using the analysis result Is to provide a method.

The execution cost analysis system according to claim 1, a transaction information storage unit that stores transaction information related to a bond trading transaction for each transaction, a data storage unit that stores a bond market price of a bond that is a target of the trading transaction, Execution cost calculation that calculates each execution cost of the trading transaction by an implementation shortfall method using the transaction information stored in the transaction information storage unit and the bond market price stored in the data storage unit and parts, and the execution cost calculation execution costs storage unit that stores the execution costs calculated by unit, and futures information storage unit for storing forward information including information indicating the price fluctuation during day bond futures, the futures The average price of the bond future is calculated using the futures information stored in the information storage unit, and the calculated average price of the bond future is calculated. Using the bond market price stored in the data storage unit and the futures information stored in the futures information storage unit, a day average price of bond actual is calculated, and the calculated bond actual price a timing cost calculation unit that calculates a timing cost of each of the transactions by calculating the difference between before Hibiki value price during the day and stored in the data storage unit, is calculated by the timing cost calculation unit A timing cost storage unit that stores the timing cost; and each timing cost of the sales transaction stored in the timing cost storage unit from each execution cost of the sales transaction stored in the execution cost storage unit calculating a dealer spread of each of the transactions by reduction dealers spread calculator A dealer spread storage unit for storing the dealer spread calculated by the dealer spread calculation unit, and each dealer spread of the sales transaction stored in the dealer spread storage unit is classified based on a type and a remaining period. A classification unit, a classification storage unit that stores the dealer spread classified by the classification unit, an average value calculation unit that calculates an average value of the dealer spread stored in the classification storage unit, and the average value calculation An average value storage unit that stores the average value calculated by the unit for each type and remaining period, and an output unit that outputs the average value for each type and remaining period stored in the average value storage unit. Features.

According to the execution cost analysis system of the first aspect, the execution cost and the timing cost relating to the bond transaction are calculated, and the dealer spread is calculated based on the execution cost and the timing cost. The average value of the calculated dealer spread is calculated for each type of bond and the remaining period. That is, the cost obtained by removing the execution cost from the timing cost that occurs when the contract price of the bond fluctuates based on fluctuations in interest rates, etc. is calculated as a dealer spread that is a commission paid to the securities company. Therefore, the execution cost, which tends to be unclear because it is executed by the relative transaction, can be broken down into two elements, the timing cost and the dealer spread, and the execution cost can be analyzed accurately. In addition, the average value of the dealer spread is calculated for each type of bond and the remaining period of the bond. Therefore, by calculating the dealer spread of the bond of the issue subject to analysis and referring to the average value of the type of the bond to which the issue of the issue belongs and the average value of the dealer spread for the remaining period, It is possible to analyze whether the dealer spread is within a reasonable range, and to analyze the execution cost in detail and accurately.
In addition, the execution cost is reduced by the implementation shortfall method, which considers the deterioration of bond prices that occur based on the fluctuations in interest rates during the period from the decision to buy and sell bonds to the actual execution of the bonds as costs. Calculated. Therefore, since the timing cost and dealer spread are included in the calculated execution cost, it is possible to calculate the dealer spread with high accuracy by excluding the timing cost from the execution cost and accurately analyze the execution cost. it can.
In addition, the timing cost generated in the buying and selling transaction of bonds (bond bonds) is calculated based on the average intraday price of bond futures. In other words, information indicating the intraday price fluctuation of the bond bond is not disclosed, but since the bond futures are traded in the open market, information indicating the price fluctuation during the day is generally disclosed. Since fluctuations in the price of futures are linked, the timing cost of the bond physical trading is calculated based on the average intraday price of bond futures. Therefore, it is possible to calculate an appropriate timing cost and analyze the execution cost appropriately for a bond transaction in which it is unknown at what time of day the transaction was performed.

The execution cost analysis system according to claim 2, wherein the futures information includes price fluctuations during the day including a price at the start of a transaction of the bond futures, a maximum price during the day, a low price during the day, and a price at the end of the transaction. It is the information which shows .

The execution cost analysis system according to claim 3 is a comparison unit that compares an average value of a dealer spread stored in the average value storage unit with a dealer spread in an arbitrary sales transaction, and a comparison in the comparison unit. The apparatus further includes a dealer spread evaluation unit that performs evaluation on the dealer spread of the arbitrary sales transaction using the result.

According to the execution cost analysis system of claim 3 , the average value of the dealer spread of the bond of the issue to be evaluated and the dealer spread of the type and remaining period to which the bond of the issue to be evaluated belongs. And determine whether the dealer spread in any trading transaction is appropriate based on the comparison result. Therefore, by referring to the analysis result of the execution cost, it is possible to accurately evaluate the validity of the dealer spread in each transaction.

An execution cost analysis program according to claim 4 is a transaction information storage function for storing a computer for analyzing execution costs in a bond trading transaction, a transaction information storage function for storing transaction information for each bond trading transaction, and the trading transaction. The data storage function for storing the bond market price of the bond that is the target of the transaction , the transaction information stored by the transaction information storage function, and the bond market price stored by the data storage function, the implementation shortfall method Futures information including an execution cost calculation function for calculating the execution cost of each trading transaction, an execution cost storage function for storing the execution cost calculated by the execution cost calculation function, and information indicating a daytime price fluctuation of bond futures The futures information storage function for storing the future information stored by the futures information storage function. Calculate the average price of bond futures during the day, calculate the average price of bond futures during the day, the bond market price stored by the data storage function, and the future information storage function By calculating the average price of the bond spot using the futures information, and calculating the difference between the calculated average price of the bond spot and the previous day discount price stored by the data storage function. A timing cost calculation function for calculating each timing cost of a sales transaction, a timing cost storage function for storing the timing cost calculated by the timing cost calculation function, and each of the sales transactions stored by the execution cost storage function reducing the timing cost of each of the transactions from the execution costs are stored by the timing cost storage function Dealers spread calculation function of calculating a dealer spread of each of the transactions by, said calculated by the dealer spread calculation function dealers spread storage function to store dealers spread, the trading stored by the dealer spreads storage function A classification function for classifying each dealer spread of a transaction based on the type and remaining period, a classification storage function for storing the dealer spread classified by the classification function, and an average value of the dealer spread stored by the classification storage function Average value calculation function for calculating the average value, average value storage function for storing the average value calculated by the average value calculation function for each type and remaining period, average value for each type and remaining period stored by the average value storage function Output It is made to function as a function.

According to the execution cost analysis program of the fourth aspect, the execution cost and the timing cost relating to the bond trading transaction are calculated, and the dealer spread is calculated based on the execution cost and the timing cost. Then, the calculated dealer spread is classified by bond type and remaining period, and an average value is calculated. Therefore, it is possible to analyze the execution cost accurately by decomposing the execution cost in the buying and selling transaction of bonds with less public information into two elements of timing cost and dealer spread. Also, since the average value of dealer spread is calculated for each bond type and bond remaining period, for example, the dealer spread of the bond of the issue being analyzed is calculated, and the issue being analyzed It is possible to perform a detailed analysis on whether or not the dealer spread of the sales transaction is within an appropriate range with reference to the average value of the dealer spread to which the bonds belong and the remaining period.
In addition, the execution cost is calculated based on the implementation shortfall method, which considers the deterioration of the bond price that occurs between the decision-making time of buying and selling and the actual trading as the cost. Therefore, since the timing cost and dealer spread are included in the calculated execution cost, it is possible to calculate the dealer spread with high accuracy by excluding the timing cost from the execution cost and accurately analyze the execution cost. it can.
In addition, the timing cost in the bond trading transaction is calculated based on the average price of bond futures during the day. That is, the price movement of the bond spot is estimated based on the price movement of the bond future, and the timing cost of the bond spot is calculated. Therefore, in general, by calculating the timing cost based on the price movement of bond futures linked to price fluctuations, it is possible to calculate an appropriate timing cost and to accurately analyze the execution cost.

The execution cost analysis program according to claim 5 , wherein the futures information includes a price at the start of the transaction of the bond futures, a maximum price during the day, a lowest price during the day, and a price at the end of the transaction. It is the information which shows price movement, It is characterized by the above-mentioned.

The execution cost analysis program according to claim 6 , wherein the computer compares the average value of dealer spreads stored in the average value storage function with dealer spreads in arbitrary sales transactions, the comparison It is further characterized by further functioning as a dealer spread evaluation function for evaluating the dealer spread of the arbitrary sales transaction using the comparison result by function.

According to the execution cost analysis program according to claim 6 , the average value of the dealer spread of the bond of the issue to be evaluated and the dealer spread of the type and remaining period of the bond of the issue to be evaluated Based on the comparison result, it is determined whether or not the dealer spread in any trading transaction is appropriate. Therefore, by referring to the analysis result of the execution cost, it is possible to accurately evaluate the validity of the dealer spread in each transaction.

The execution cost analysis method according to claim 7 is a transaction information storage step for storing transaction information relating to a bond sales transaction for each transaction, and a data storage step for storing a bond market price of the bond that is the target of the transaction. And execution cost calculation for calculating each execution cost of the trading transaction by an implementation shortfall method using the transaction information stored in the transaction information storage step and the bond market price stored in the data storage step An execution cost storage step for storing the execution cost calculated in the execution cost calculation step, a futures information storage step for storing futures information including information indicating price fluctuations of bond futures during the day, and the futures The date of the bond future using the futures information stored in the information storage step The average price of the bond future is calculated, and the bond market price stored in the data storage step and the future information stored in the future information storage step are used to calculate the bond price. Calculate the average price of the physical spot during the day, and calculate the difference between the calculated average price of the bond spot during the day and the previous day discount price stored in the data storage step to determine the timing of each trading transaction. a timing cost calculation step of calculating the costs, the timing cost storage step for storing the timing cost calculated in the timing cost calculation step, from said execution costs of each of the stored the sales transaction in the execution costs storage step the trading stored in the timing cost storage step And dealer spread calculating a dealer spread of each of the transactions by reduction of each of the timing cost of arguments, and dealer spreads storage step of storing said dealer spread calculated in the dealer spread calculating step, wherein A classification step for classifying each dealer spread of the sales transaction stored in the dealer spread storage step based on a type and a remaining period, a classification storage step for storing the dealer spread classified by the classification step, and the classification An average value calculating step for calculating an average value of the dealer spread stored in the storing step, and an average value calculated in the average value calculating step are stored for each type and remaining period. An average value storing step; and an output step of outputting an average value for each type and remaining period stored in the average value storing step .

According to the execution cost analysis method of the seventh aspect, the execution cost and the timing cost relating to the bond trading transaction are calculated, and the dealer spread is calculated based on the execution cost and the timing cost. Then, the calculated dealer spread is classified by bond type and remaining period, and an average value is calculated. Therefore, the execution cost, which tends to be unclear because it is executed by the relative transaction, is broken down into two elements, the timing cost and the dealer spread. Analysis can be performed easily. In addition, the average value of the dealer spread is calculated for each type of bond and the remaining period of the bond. Therefore, by calculating the dealer spread of the bond of the issue subject to analysis and referring to the average value of the type of the bond to which the issue of the issue belongs and the average value of the dealer spread for the remaining period, It is also possible to analyze whether or not the dealer spread is within an appropriate range, and the execution cost can be analyzed in detail and accurately.
In addition, the execution cost is calculated based on the implementation shortfall method, which considers the deterioration of the bond price that occurs between the decision-making time of buying and selling and the actual trading as the cost. Therefore, since the timing cost and dealer spread are included in the calculated execution cost, it is possible to calculate the dealer spread with high accuracy by excluding the timing cost from the execution cost and accurately analyze the execution cost. it can.
In addition, the timing cost in the bond trading transaction is calculated based on the average price of bond futures during the day. Therefore, since the price fluctuations are generally linked to bond actuals and bond futures, it is possible to calculate a reasonable timing cost and to accurately analyze the execution cost.

Further, the execution cost analysis method according to claim 8 , wherein the futures information includes a price at the start of the transaction of the bond futures, a maximum price during the day, a low price during the day, and a price at the end of the transaction. It is the information which shows price movement, It is characterized by the above-mentioned.

The execution cost analysis method according to claim 9 is a method of comparing the average value of the dealer spread with a dealer spread in an arbitrary sales transaction and using the comparison result in the comparison step, A dealer spread evaluation step for evaluating the dealer spread of the transaction.

According to the execution cost analysis method described in claim 9 , the average value of the dealer spread of the bond of the issue to be evaluated and the dealer spread of the type and remaining period to which the bond of the issue to be evaluated belongs. Based on the comparison result, it is determined whether or not the dealer spread in any trading transaction is appropriate. Therefore, by referring to the analysis result of the execution cost, it is possible to accurately evaluate the validity of the dealer spread in each transaction.

  According to the execution cost analysis system according to the present invention, it is a fee paid to a securities company, which is a cost obtained by excluding a timing cost generated by a change in the price of a bond from an execution cost based on a change in interest rates, etc. Calculated as dealer spread. Therefore, the execution cost related to the buying and selling transactions of bonds that are carried out by relative transactions and for which information relating to selling and selling transactions is not disclosed is broken down into two elements: timing costs and dealer spreads. It is possible to accurately analyze whether or not. In addition, since the average value of dealer spread is calculated for each type of bond and the remaining period, the dealer spread of the bond of the issue subject to analysis is calculated, and the bond of the issue of the issue belongs Execution costs can be analyzed in detail with reference to average dealer spreads of type and remaining period. Furthermore, the average value of the dealer spread for each bond type and remaining period is used as a useful judgment material when, for example, determining whether the dealer spread set by the company is at an appropriate level in a securities company. be able to.

  Further, the validity of the dealer spread can be evaluated by comparing the average value of the dealer spread, which is the analysis result of the execution cost, with the dealer spread in an arbitrary sales transaction. Therefore, for example, the evaluation result relating to the validity of the dealer spread included in the execution cost can be used as useful data when selecting a securities company that requests a sales transaction.

  Further, according to the execution cost analysis program according to the present invention, the execution cost analysis program is incorporated into the computer, whereby the computer can be functioned as an execution cost analysis system for analyzing the execution cost in the buying and selling transaction of bonds. Therefore, by incorporating the execution cost analysis program into the computer, it is possible to easily analyze the execution cost in bond trading and evaluate the validity of the dealer spread included in the execution cost.

  Further, according to the execution cost analysis method according to the present invention, even if it is difficult to obtain information indicating the price fluctuation of the bond and the detailed information regarding the bond transaction is unknown, The execution cost can be analyzed accurately. Therefore, it is possible to easily analyze whether or not the price presented by the securities company in the relative transaction is an appropriate price.

  Hereinafter, an execution cost analysis system according to an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a block configuration diagram of the entire system including the execution cost analysis system 2. As shown in FIG. 1, the execution cost analysis system 2 is connected via a network 4 to a stock exchange system 6 installed in a stock exchange such as the Tokyo Stock Exchange. Further, the execution cost analysis system 2 is connected to operation company systems 8a, 8b, and 8c installed in each of a plurality of operation companies via a network 4.

  FIG. 2 is a block configuration diagram of the execution cost analysis system 2 according to the embodiment of the present invention. As shown in FIG. 2, the execution cost analysis system 2 includes a data control unit 10 that controls processing of data in the execution cost analysis system 2. The data control unit 10 includes a communication control unit 12 that controls communication between the stock exchange system 6 and the management company systems 8a, 8b, and 8c, and a relative transaction between the management company and the securities company. A transaction information storage unit 14 for storing transaction information related to bond sales transactions is connected. In addition, the data control unit 10 stores a futures information storage unit 16 that stores futures information related to price movements of bond futures that are traded in the securities trading market, and reference prices (bond prices) of bonds such as trading reference statistics. A data storage unit 18 for storage is connected. The data control unit 10 is connected to an output unit 20 that outputs an execution cost analysis result and the like. The output unit 20 includes a display device such as a liquid crystal display and a printing device such as a printer.

  In the transaction information storage unit 14, bond transaction information transmitted from each of the management company systems 8 a, 8 b, 8 c via the network 4 and received via the communication control unit 12, that is, a bond made by the management company Is stored for each transaction. This transaction information includes information such as the brand to be traded, transaction type (type of selling or buying), contract amount and contract unit price, etc., for example, once a day for an operating company system every predetermined period. Each of 8a, 8b, and 8c is transmitted to the execution cost analysis system 2. In the execution cost analysis system 2, transaction information is received via the communication control unit 12, and the received transaction information is stored in the transaction information storage unit 14. The transaction information storage unit 14 stores, for example, transaction information related to bond trading transactions for each day in the past year transmitted from the management company systems 8a, 8b, and 8c.

  Further, information (maturity date information) indicating the maturity date of the bond on which the transaction has been performed is transmitted together with the transaction information from each of the management company systems 8a, 8b, and 8c. The execution cost analysis system 2 receives the maturity date information together with the transaction information via the communication control unit 12 and stores the maturity date information together with the transaction information in the transaction information storage unit 14. Note that the remaining period of the bond for which the transaction was performed is calculated based on the received maturity date information, and is stored in the data storage unit 14 together with the transaction information and the maturity date information.

  The futures information storage unit 16 stores futures information on bond futures transmitted from the stock exchange system 6 via the network 4 and received via the communication control unit 12. This futures information includes information indicating the intraday price fluctuation of each stock of bond futures on the stock exchange, such as information indicating the price of bond futures at each time (tick-by-tick price information), bond Information such as the price of futures at the start of trading, the highest price during the day, the lowest price during the day, and the price (closing price or discount price) at the end of the trading are included. The futures information storage unit 16 stores, for example, futures information indicating price movements of bond futures for each day in the past year.

  Further, the data storage unit 18 is configured by bond market prices of each bond issue that is a transaction target, for example, trading reference statistical values published by the Japan Securities Dealers Association, arbitrary securities companies, financial institutions, and the like. Information indicating the bond market price announced as a reference price by an organization or the like is stored. Here, trading reference statistical values are prices and yields of bonds that are used as reference for trading in the public and corporate bond over-the-counter market, and are calculated and published by the Japan Securities Dealers Association. In other words, at the Japan Securities Dealers Association, as the price of each bond issue at 15:00 on the same day (the day before the date of announcement of trading reference statistics) from multiple securities companies, an intermediate price between the selling price and the buying price ( Value) is reported. Then, an average value at 15:00 is calculated for each brand based on the median values reported from a plurality of securities companies, and the calculated average value is published for each brand as a trading reference statistical value.

  FIG. 3 is a block diagram of the operating company system 8a according to the embodiment of the present invention. As shown in FIG. 3, the operating company system 8a includes a data control unit 30 that controls data processing in the operating company system 8a. A communication control unit 32 that controls communication with the execution cost analysis system 2 is connected to the data control unit 30. Further, the data control unit 30 is connected to a data storage unit 34 that stores transaction information and the like related to bond trading transactions performed by the asset management company.

  In the data storage unit 34, transaction information related to a bond trading transaction performed by an asset management company, that is, information including a trading target brand, a transaction type, a contract amount, a contract unit price, etc., and information indicating a bond maturity date (maturity date) Information) is stored. The transaction information and maturity date information stored in the data storage unit 34 are transmitted to the execution cost analysis system 2 via the communication control unit 32 once a day, for example, every predetermined period. . Note that the configuration of the operating company systems 8b and 8c is the same as the configuration of the operating company system 8a, and a description thereof will be omitted. Also in the management company systems 8b and 8c, transaction information related to bond trading transactions performed by the management company at predetermined intervals is transmitted to the execution cost analysis system 2 via the network 4.

  Next, execution cost analysis processing by the execution cost analysis system according to the embodiment of the present invention will be described with reference to the flowchart of FIG. In addition, the process demonstrated below is performed in the data control part 10 of the execution cost analysis system 2. FIG.

  First, based on the transaction information stored in the transaction information storage unit 14, the execution cost in the bond trading transaction is calculated (step S10). That is, the execution cost for each transaction is calculated based on the implementation shortfall (IS) method. Here, the IS method means that the market price fluctuates due to the influence of interest rates, etc. in the process from the time when a decision to buy or sell is made until the bond is actually purchased (sold), and the cost equivalent to the commission by the securities company. This is a method that considers price deterioration caused by imposition as a cost. In order to use this IS method, first, the contract unit price of the bond in the transaction is extracted from the transaction information for each transaction stored in the transaction information storage unit 14. Next, the bond market price on the day before the date of the transaction is extracted from the bond market price stored in the data storage unit 18. Then, the execution cost of the transaction is calculated by applying the contracted unit price extracted from the transaction information and the extracted bond market price to the following (Formula 1).

Execution cost = (P ₋₁ −e) bs_flag (Formula 1)
In this (Formula 1), P ₋₁ is the bond market price of the bond on the day before the transaction date, and e is the contract unit price of the bond in the transaction. The bs_flag is a coefficient indicating a flag that is “1” when a bond is purchased and “−1” when the bond is sold. Here, since the amount of the bond is determined by the interest rate of the bond, the value of the execution cost of the bond may be indicated by the yield. The same applies to all the above costs. The calculated execution cost is stored in the transaction information storage unit 14 in correspondence with the transaction information.

Here, in step S10, since the execution cost is calculated by the IS method, the calculated execution cost includes a timing cost and a dealer spread.
Therefore, the dealer spread value can be calculated with high accuracy by excluding the timing cost from the execution cost. Note that the timing cost is the cost that is incurred when the bond price fluctuates due to interest rate fluctuations, etc., compared to the previous day. . Dealer spread is the cost that indicates the commission to the dealer (securities company). The dealer spreads to the customer (management company) reflecting the interest rate level at the time of the transaction, the type of bond, the remaining period, the transaction scale, etc. It is indicated by the divergence between the offered trading price and the bond price at that time.

  FIG. 5 is a diagram schematically showing the concept of bond execution costs according to the embodiment of the present invention. As shown in FIG. 5, the execution cost includes a timing cost and a dealer spread. That is, as shown in FIG. 5, the discount price (closing price) of the previous day, that is, the bond market price of the previous day and the bond market price of the current day are connected by a straight line, and the horizontal axis represents time and the vertical axis represents price. Suppose that the transition of the middle value of a bond is shown. In this case, the difference between the bond price indicated by the graph of FIG. 5 and the contract price of the bond at any time during the day is the dealer spread. Further, as shown in FIG. 5, the timing cost is the difference between the bond market price shown at the “CLOSE” time of the previous day and the midpoint value at the time of execution. The sum of the dealer spread and the timing cost is the execution cost of the sales transaction.

  Next, a timing cost included in the execution cost calculated in step S10 is calculated (step S11). Here, for example, as shown in FIG. 5, if the bond market is rising, buying late will cause the timing cost to increase, and conversely if the bond market is falling, buying late Doing so is a factor that lowers timing costs. However, the transaction information stored in the transaction information storage unit 14 may not include the time at which the sales transaction is actually performed. In such a case, the transaction transaction is performed at what time of day. It is unknown whether it was done. The data storage unit 18 stores the bond market price for each day, but does not store information indicating the price fluctuation during the day. On the other hand, for bond futures that are traded in the market, futures information including information indicating price fluctuations during the day is stored in the futures information storage unit 16. That is, the futures information stored in the futures information storage unit 16 includes information indicating price movements during the day, such as the price at the start of trading, the highest price during the day, and the lowest price during the day. Changes in bond futures prices are linked. Therefore, the average daytime price of bond futures is calculated based on futures information stored in the futures information storage unit 16, and the average daytime price of bond futures is calculated as a timing cost calculation target. Convert to the average price of the physical day. Hereinafter, an example of processing for calculating the timing cost of the bond actual will be shown.

First, from the futures information stored in the futures information storage unit 16, the futures information of the bond futures that are traded on the same day as the bond actual trading is extracted. Then, based on the extracted forward information is calculated on the basis of the showing the average price FP _a daytime bond futures below (Equation 2).
FP _a = (FP _o + FP _h + FP ₁ + FP _C ) / 4 (Expression 2)
Here, FP _o is the price of the bond future at the start of trading on that day, and FP _h is the highest price of the bond future on that day. FP _l is the lowest price of the bond future on that day, and FP _c is the price (discount or closing price) of the bond future at the end of the transaction on that day.

Next, the price fluctuation range of bond futures during the day is calculated from the midpoint of the line connecting the previous day discount price FP _-1 and the current day (the day on which the bond futures transaction is conducted) FP _c in a straight line. _Let us characterize the distance to the average price FP _a . That is, as shown in FIG. 6A, than the price {(FP _-1 + FP _c ) / 2} of the midpoint of the line segment connecting the previous discount FP _-1 and the current discount FP _c of bond futures with a straight line. It is considered that the daytime average price FP _a calculated by (Formula 2) is higher means that the daytime price fluctuation was above the straight line. Here, since the price fluctuation of bond futures diverges above the straight line, as shown in FIG. 6B, the bond market price P _{−1 of the} previous day is similar to the price fluctuation of bond futures as shown in FIG. If prices than the midpoint of the line segment connecting the current day of the bond market P _c by a straight line _{{(P -1 + P c)} / 2}, the average price P _a becomes so price fluctuates high intraday Can be estimated. In addition, the coefficient indicating the price fluctuation range of bond futures, that is, the difference between the midpoint price {(FP ₋₁ + FP _c ) / 2} and the average price FP _a during the day, that is, price variability σ _FP (volatility) assuming proportional to the value, the volatility indicating the price volatility of the bond spot and sigma _P, can be calculated an average price P _a daytime bond kind based on the following (equation 3).

上述の（数式３）に基づいて算出された債券現物の日中の平均価格Ｐ_ａは、データ記憶部１８に記憶される。このデータ記憶部１８に記憶された債券現物の日中の平均価格Ｐ_ａとデータ記憶部１８に記憶されている前日引値ＦＰ_−１との差を算出する。そして、算出された差分が、執行コストに含まれるタイミングコストとして取引情報に対応させて取引情報記憶部１４に記憶される。

Price P _a daytime bonds actual calculated based on the above (Equation 3) is stored in the data storage unit 18. Calculating the difference between the pre Hibiki value FP _-1 stored in the price P _a and the data storage unit 18 in the day stored bond kind in the data storage unit 18. The calculated difference is stored in the transaction information storage unit 14 in correspondence with the transaction information as a timing cost included in the execution cost.

  Next, a dealer spread included in the execution cost calculated in step S10 is calculated (step S12). That is, the dealer spread is calculated by subtracting the timing cost calculated in step S11 from the execution cost calculated in step S10, and stored in the transaction information storage unit 14.

  Next, the dealer spread stored in the transaction information storage unit 14 is classified based on the bond type and the remaining period (step S13). For example, dealer spreads are classified according to the types and remaining periods (years) of “0”, “1”, “2”,. Note that dealer spreads classified by bond type and remaining period are stored in the data storage unit 18.

  Next, the dealer spread values classified in step S13 are aggregated (step S14), and an average value of dealer spreads (average dealer spread) is calculated for each bond type and remaining period based on the aggregated values. (Step S15).

  The calculated average dealer spread is stored in the data storage unit 18 for each bond type and remaining period (step S16), and the stored average dealer spread is output via the output unit 20 (step S17). .

  FIG. 7 is a diagram showing an example of an average value dealer spread according to the embodiment of the present invention. As shown in FIG. 7, the average dealer spread is 0-15 for each type of bond: yen-denominated foreign bonds, discounted government bonds, financial bonds, business bonds, government-guaranteed bonds, municipal bonds, power bonds, special bonds, and interest-bearing bonds. Calculated for each remaining year. Further, as shown in FIG. 7, an average value dealer spread for each remaining period, for example, an average value dealer spread “0.005%” for all types when the remaining period is “0” is shown. Further, an average dealer spread for each type, for example, an average dealer spread “0.0034%” over the entire period for yen-denominated bonds is shown.

  Note that the average dealer spread shown in FIG. 7 is updated every predetermined period, for example, every month. That is, the transaction information storage unit 14 stores daily transaction information transmitted from each of the operating company systems 8a, 8b, and 8c via the network 4. Therefore, for example, the average dealer spread is calculated based on the transaction information of transactions on each day for one year from October 2002 to September 2003, and the transaction information on each day of October 2003 is newly added. It is assumed that it is stored in the transaction information storage unit 14. In this case, for example, based on the transaction information for each day from October 2002 to October 2003, or for example, based on the transaction information for each day from November 2002 to October 2003. The value dealer spread is calculated, and the average value dealer spread is updated.

  Next, processing performed by the execution cost analysis system according to the embodiment will be described with reference to the flowchart of FIG. In the following, processing for evaluating execution costs in trading transactions of arbitrary bonds using the average dealer spread calculated for each bond type and remaining period will be described.

  First, the execution cost of the transaction to be evaluated is calculated (step S20), and the timing cost included in the calculated execution cost is calculated (step S21). Then, a dealer spread is calculated based on the calculated execution cost and timing cost (step S22), and the calculated dealer spread is stored in the transaction information storage unit 14 in association with the transaction information. In addition, since the process of step S20-step S22 is the same process as the process in step S10-step S12 of the flowchart shown in FIG. 4, detailed description is abbreviate | omitted.

  Next, the calculated dealer spread is compared with the average dealer spread stored in the data storage unit 18 (step S23). That is, the dealer spread of the bond of the brand to be evaluated is compared with the average dealer spread of the type and remaining period to which the bond of the brand to be evaluated belongs. For example, if the transaction being evaluated is a business bond with a remaining period (year) of “10”, the remaining period becomes “10” from the average dealer spread shown in FIG. The average dealer spread “0.080%” of the existing corporate bonds is extracted. Then, for example, the value of the dealer spread of the transaction to be evaluated is compared with the extracted average dealer spread “0.080%”.

  Next, the dealer spread is evaluated based on the comparison result in step S23 (step S24). For example, when the dealer spread value of the transaction to be evaluated is a value of “0.072% or less” with respect to the average dealer spread “0.080%”, “good”, “0 0.073% or more and 0.088% or less ”is“ appropriate ”, and“ 0.089% or more and 1.04% or less ”is“ standard ”or“ 1.05 ”. If the value is “% or more”, an evaluation such as “inappropriate” is performed. Note that a threshold value for evaluating whether the value of the dealer spread is appropriate with respect to the value of the average value dealer spread can be arbitrarily set. Further, the evaluation result is stored in the transaction information storage unit 14 in association with the transaction information regarding the transaction to be evaluated.

  Next, the evaluation result memorize | stored in the transaction information storage part 14 is output via the output part 20 (step S25). That is, the evaluation result stored in the transaction information storage unit 14 is output by printing or the like via the output unit 20 configured by a printing device such as a printer. The output evaluation results are reported to the asset management company. In addition, for example, when an execution cost analysis result including an evaluation result of a dealer spread is created, the created execution cost analysis result is managed by an operating company system installed in an investment company every predetermined period, for example, an operation By transmitting to the company system 8a via a network, the evaluation result may be output and reported to the operating company.

  According to the execution cost analysis system according to the embodiment of the present invention, the execution cost in each transaction is calculated based on the transaction information related to the bond buying and selling transaction transmitted from each of the plurality of asset management companies. The dealer spread is calculated by decomposing the execution cost into two elements: timing cost and dealer spread. Then, the dealer spread is classified for each type of bond and the remaining period, and an average dealer spread for each type and remaining period is calculated and output. Accordingly, it is possible to accurately grasp the ratio of the dealer spread to the execution cost in the bond buying and selling transaction performed in the relative transaction, and to accurately analyze the execution cost.

  In addition, according to the execution cost analysis system according to the embodiment of the present invention, the execution cost in the buying and selling transaction of bonds is calculated using the IS method. Therefore, since the execution cost includes the timing cost and the dealer spread, the dealer spread value can be estimated with high accuracy by excluding the timing cost from the execution cost. Therefore, it is possible to accurately analyze the extent of the dealer spread corresponding to the fee paid to the dealer (securities company).

  In addition, according to the execution cost analysis system according to the embodiment of the present invention, since the price fluctuations of the bond spot and the bond future are generally linked, it becomes an execution cost analysis object from the price fluctuation of the bond future during the day. The timing cost is calculated by estimating the intraday price fluctuation of the bonds. Therefore, since it is possible to calculate a reasonable timing cost for a bond transaction in which it is unknown at what time of day the trading was executed, the execution cost can be analyzed appropriately.

  Further, according to the execution cost analysis system according to the embodiment of the present invention, the calculated dealer spread is classified for each type of bond and the remaining period, and an average dealer spread is calculated for each type and remaining period. . Accordingly, it is possible to analyze in detail the execution cost that fluctuates in consideration of the risk of trading transactions based on the type of bond and the remaining period.

  In addition, according to the execution cost analysis system according to the embodiment of the present invention, the dealer spread of the bond of the issue to be evaluated, the type to which the bond of the issue to be evaluated belongs and the average of the remaining period By comparing with the value dealer spread, it is possible to evaluate whether or not the dealer spread to be evaluated is appropriate. Therefore, it is possible to confirm whether or not the fee paid to the securities company is within an appropriate range, and to use the evaluation result as useful data when considering more efficient asset management, for example. In addition, since the average dealer spread is output for each type of bond and the remaining period, the validity of the dealer spread in the sales transaction can be analyzed accurately and in detail.

  In the above-described embodiment, the execution cost analysis system is described. However, by incorporating the execution cost analysis program into the computer, the computer is configured as the same system as the execution cost analysis system according to the above-described embodiment. Can function. Therefore, the computer in which the execution cost analysis program is incorporated may perform the processing in the above-described embodiment.

  Further, in the execution cost analysis system according to the above-described embodiment, the average price of bond bonds in the daytime is estimated based on the average of four values of bond futures, but the VWAP (Volume Weighted) of bond futures You may make it estimate the average daytime price of the bond thing based on Average Price (trading weight weighted average price). That is, when the futures information includes the price information of tick-by-ticks of bond futures, VWAP is calculated based on the price information of the tick-by-ticks, and VWAP is shown in FIG. The daytime average price of the bond in question may be estimated by using the difference from the midpoint of the graph shown as the price change range.

  Further, in the execution cost analysis system according to the above-described embodiment, in (Equation 3), it is defined that the price difference is proportional to the volatility, but it may be defined that the price change rate is proportional to the volatility. . Also, instead of volatility, we define that the rate of price change is proportional to the duration (a coefficient indicating how much the price of a bond changes based on fluctuations in interest rates), and estimate the average price of the bond in the daytime. May be.

  Further, in the execution cost analysis system according to the above-described embodiment, the difference between the daytime average price and the contract price is estimated as the timing cost, but the process for calculating the timing cost may be changed depending on the transaction time. . That is, the timing cost is a cost that should be taken into account when contracted at various times during the day or when the contracted time is not clear. Therefore, when the transaction time is clear, for example, when the contract is executed in the morning based on the previous day's discount, the timing cost is set to “0”, and when the contract is executed based on the current day's discount, the timing cost is set to the previous day. The timing cost may be calculated by a different process according to the contract time, such as a price difference from the discount price to the current day discount price.

  Further, in the execution cost analysis system according to the above-described embodiment, the evaluation result is output, and the output evaluation result is reported to the management company, but the average for each bond type and remaining period shown in FIG. You may make it provide a value dealer spread to an operation company. In this case, for example, the average dealer spread can be referred to as a useful material when examining whether or not asset management is appropriately performed in the management company. Moreover, you may make it provide an average value dealer spread to a securities company. In this case, the securities company can refer to the calculated average dealer spread and analyze whether or not its dealer spread is appropriate.

  In the execution cost analysis system according to the above-described embodiment, the average dealer spread is updated every month. For example, the average value every arbitrary period such as once every quarter or once every half period. The dealer spread may be updated. Further, for example, every time execution costs are analyzed based on a request from an investment company, an average is calculated based on transaction information in a predetermined period including the latest transaction information stored in the transaction information storage unit 14 at that time. A value dealer spread may be calculated.

  In addition, in the execution cost analysis system according to the above-described embodiment, the management company is provided with evaluation results of dealer spreads included in the execution cost in arbitrary transactions. You may make it provide the management company the evaluation result regarding a condition. For example, based on transaction information related to bond trading transactions conducted by a given manager for a certain period, create bond manager average dealer spreads classified by bond type and remaining period as shown in Fig. 7. To do. Then, the created management company average dealer spread and the average dealer spread (see FIG. 7) for each bond type and remaining period calculated based on transaction information of a plurality of management companies are provided to the management company. You may do it. In this case, the management company compares the average dealer spread with the average dealer spread and the average dealer spread for each bond type and the remaining period, Evaluation can be made. For example, regarding bond sales transactions conducted by the asset management company, all bonds are traded appropriately, such as which bonds are properly traded and no bonds are traded properly. It is possible to accurately analyze and understand the execution cost of.

  In the above-described embodiment, the maturity date information is transmitted from the management company system, but the maturity date information may not be transmitted from the management company system. Therefore, when the maturity date information is not transmitted from the operating company system, the maturity date information may be acquired separately by the execution cost analysis system. That is, the execution cost analysis system is equipped with a database that stores information on bonds, in which the maturity date information of bonds is searched based on bond issues, and the found maturity date information is stored in the transaction information storage unit together with the transaction information. You may make it do.

It is a block block diagram of the whole system containing the execution cost analysis system which concerns on embodiment of this invention. It is a block block diagram of the execution cost analysis system which concerns on embodiment of this invention. It is a block block diagram of the operation company system which concerns on embodiment of this invention. It is a flowchart for demonstrating the analysis process of the execution cost by the execution cost analysis system which concerns on embodiment of this invention. It is the figure which showed typically the concept of the execution cost which concerns on embodiment of this invention. It is a figure for demonstrating the process which calculates the timing cost which concerns on embodiment of this invention. It is a figure which shows an example of the dealer spread which concerns on embodiment of this invention. It is a flowchart for demonstrating the process which evaluates the dealer spread by the execution cost analysis system which concerns on embodiment of this invention.

Explanation of symbols

2 ... Execution cost analysis system, 4 ... Network, 6 ... Stock exchange system, 8a, 8b, 8c ... Management company system, 10 ... Data control unit, 12 ... Communication control unit, 14 ... Transaction information storage unit, 16 ... Futures Information storage unit, 18 ... data storage unit, 20 ... output unit, 30 ... data control unit, 32 ... communication control unit, 34 ... data storage unit.

Claims (9)

A transaction information storage unit for storing transaction information related to bond sales transactions for each transaction;
A data storage unit for storing the bond market price of the bond that is the subject of the transaction;
Execution cost calculation that calculates each execution cost of the trading transaction by an implementation shortfall method using the transaction information stored in the transaction information storage unit and the bond market price stored in the data storage unit and parts,
An execution cost storage unit for storing the execution cost calculated by the execution cost calculation unit;
A futures information storage unit for storing futures information including information indicating price fluctuations of bond futures during the day;
A daytime average price of the bond future is calculated using the futures information stored in the futures information storage unit, and the calculated daytime average price of the bond future is stored in the data storage unit. The bond market price and the futures information stored in the futures information storage unit are used to calculate the average price of the bond spot during the day, and the calculated average price of the bond spot and the day price A timing cost calculation unit for calculating each timing cost of the buying and selling transaction by calculating a difference with the previous discount price stored in the data storage unit ;
A timing cost storage unit for storing the timing cost calculated by the timing cost calculation unit;
Each dealer spread of the sales transaction is reduced by subtracting each timing cost of the sales transaction stored in the timing cost storage unit from each execution cost of the sales transaction stored in the execution cost storage unit. and the dealer spread calculation unit for calculating a,
A dealer spread storage unit for storing the dealer spread calculated by the dealer spread calculation unit;
A classification unit that classifies each dealer spread of the sales transaction stored in the dealer spread storage unit based on a type and a remaining period;
A classification storage unit for storing the dealer spread classified by the classification unit;
An average value calculation unit for calculating an average value of the dealer spread stored in the classification storage unit;
An average value storage unit that stores the average value calculated by the average value calculation unit for each type and remaining period;
An execution cost analysis system comprising: an output unit that outputs an average value for each type and remaining period stored in the average value storage unit . The futures information is information indicating price movements during the day, including a price at the start of a transaction of the bond futures, a highest price during the day, a lowest price during the day, and a price at the end of the transaction. The execution cost analysis system according to 1. A comparison unit that compares an average value of dealer spreads stored in the average value storage unit with a dealer spread in an arbitrary sales transaction;
With comparison result of the comparison unit, according to claim 1 or claim 2 executive cost analysis system according to, further comprising a dealer spread evaluating unit for performing an evaluation of the dealer spread of the arbitrary transactions. Computers to analyze execution costs in bond trading
Transaction information storage function for storing transaction information related to bond sales transactions for each transaction,
A data storage function for storing the bond market price of the bond that is the subject of the transaction;
An execution cost calculation function for calculating each execution cost of the trading transaction by an implementation shortfall method using the transaction information stored by the transaction information storage function and the bond market price stored by the data storage function;
An execution cost storage function for storing the execution cost calculated by the execution cost calculation function;
Futures information storage function to store futures information including information indicating price fluctuations of bond futures during the day,
A daytime average price of the bond future is calculated using the futures information stored by the futures information storage function, and the calculated daytime average price of the bond future is stored by the data storage function. The bond average price and the future price information stored by the futures information storage function are used to calculate the average price of the bond spot during the day, and the calculated average price of the bond spot during the day and the data storage function. A timing cost calculation function for calculating each timing cost of the buying and selling transaction by calculating a difference from the previous discount price stored by
A timing cost storage function for storing the timing cost calculated by the timing cost calculation function;
Each dealer spread of the sales transaction is calculated by subtracting each timing cost of the sales transaction stored by the timing cost storage function from each execution cost of the sales transaction stored by the execution cost storage function Dealer spread calculation function ,
A dealer spread storage function for storing the dealer spread calculated by the dealer spread calculation function;
A classification function for classifying each dealer spread of the sales transaction stored by the dealer spread storage function based on the type and the remaining period;
A classification storage function for storing the dealer spread classified by the classification function;
An average value calculation function for calculating an average value of the dealer spread stored by the classification storage function;
An average value storage function for storing the average value calculated by the average value calculation function for each type and remaining period;
An execution cost analysis program that functions as an output function for outputting an average value for each type and remaining period stored by the average value storage function . The futures information is information indicating price movements during the day, including a price at the start of a transaction of the bond futures, a highest price during the day, a lowest price during the day, and a price at the end of the transaction. 4. Execution cost analysis program according to 4. The computer,
A comparison function for comparing an average value of dealer spreads stored in the average value storage function with a dealer spread in an arbitrary sale transaction;
6. The execution cost analysis program according to claim 4 , wherein the execution cost analysis program is further made to function as a dealer spread evaluation function for performing an evaluation on a dealer spread of the arbitrary sales transaction using a comparison result by the comparison function. A transaction information storage step for storing transaction information related to bond sales transactions for each transaction;
A data storage step of storing the bond market price of the bond that is the subject of the buying and selling transaction;
An execution cost calculation step of calculating each execution cost of the trading transaction by an implementation shortfall method using the transaction information stored in the transaction information storage step and the bond market price stored in the data storage step; ,
An execution cost storage step for storing the execution cost calculated in the execution cost calculation step;
A futures information storage step for storing futures information including information indicative of intraday price fluctuations of bond futures;
A daytime average price of the bond future is calculated using the futures information stored in the futures information storage step, and the calculated daytime average price of the bond future is stored in the data storage step. The average price of the bond spot is calculated using the bond market price and the future information stored in the future information storage step, and the calculated average price of the bond spot and the data storage step is calculated. A timing cost calculating step of calculating each timing cost of the buying and selling transaction by calculating a difference from the previous day discount price stored in
A timing cost storing step for storing the timing cost calculated in the timing cost calculating step;
Each dealer spread of the sales transaction is calculated by subtracting each timing cost of the sales transaction stored in the timing cost storage step from each execution cost of the sales transaction stored in the execution cost storage step Dealer spread calculation step ,
A dealer spread storing step for storing the dealer spread calculated in the dealer spread calculating step;
A classification step of classifying each dealer spread of the sales transaction stored in the dealer spread storage step based on a type and a remaining period;
A classification storage step for storing the dealer spread classified by the classification step;
An average value calculating step of calculating an average value of the dealer spread stored in the classification storing step;
An average value storing step for storing the average value calculated in the average value calculating step for each type and remaining period;
An execution cost analysis method comprising: an output step of outputting an average value for each type and remaining period stored in the average value storage step . The futures information is information indicating price movements during the day, including a price at the start of a transaction of the bond futures, a highest price during the day, a lowest price during the day, and a price at the end of the transaction. 7. Execution cost analysis method according to 7. A comparison step of comparing the average value of the dealer spread with a dealer spread in an arbitrary sales transaction;
The execution cost analysis method according to claim 7 or 8 , further comprising a dealer spread evaluation step of performing an evaluation on a dealer spread of the arbitrary sales transaction using the comparison result in the comparison step.

Images