KR100597862B1 - Financial Information Service for Future Return and Risk - Google Patents

Abstract

The present invention relates to a method of calculating and servicing expected returns and expected risks according to a client's portfolio in a statistical manner.

Expected return and expected risk financial information service method according to an embodiment of the present invention, calculates the daily return of stocks of each individual stock, the daily return of the industry of the industry to which the individual stock belongs and the daily market return of the market to which the individual stock belongs do. When the daily stock portfolio information is received from the client, the daily investment ratio of the client is calculated, and future expected returns and expected risks are statistically calculated and provided to the client by using bootstrap or multiple regression analysis.

According to the present invention, since the expected return and expected risk for tomorrow or the short term are provided to the client by using the historical yield data, the industry return rate and the market rate return data of the past period, the securities related website provides the user with The quality of service is improved.

Expected Return, Portfolio, Stocks, Bootstrap, Multiple Regression

Description

Expected Return and Risk Risk Financial Information Service Method {Financial Information Service for Future Return and Risk}             

1 is a block diagram showing an expected return and expected risk information service system to which the present invention is applied;

2 is a flowchart illustrating a process of calculating an expected return and an expected risk of an investor according to the present invention;

3 is an operation flowchart illustrating a process of calculating an expected return rate and an expected risk using a bootstrap method;

4 is a flowchart illustrating a process of calculating an expected return rate and an expected risk using a multiple regression analysis method.

        Explanation of symbols on the main parts of the drawings

101: web server 102: expected return / expected risk calculation server

103: database management server 104: item database

105: industry database 106: ledger database

107: Member Database 108: Rate of Return / Risk Database

111: Network 112: Korea Securities Computer System

121: Internet 122: Personal Client

123: securities company client 124: corporate client

The present invention relates to a method of serving a user by calculating, in a statistical manner, an expected return and an expected risk according to a client's portfolio using a computer.

The Internet, initially used for information sharing, is now a medium for expanding new industrial sectors, providing opportunities to move existing traditional industries into cyberspace and to create commercial added value. Recently, with the increase of securities transactions using the Internet, securities firms' websites and portal websites provide securities-related contents to their customers.

However, the securities-related content so far is only recorded on the price and details of the purchase or sale of a particular stock by the Internet user, or only the percentage of the loss or gain from the past purchase / sell if the user has bought or sold the stock. Since it is only a basic concept to inform using, there is a problem in that it is not helpful to buy / sell shares in practical and concrete.

An object of the present invention devised to solve the problems of the prior art is to provide a method of calculating an expected return and an expected risk according to a customer's portfolio to serve a corresponding customer.

Expected rate of return and expected risk financial information service method according to an embodiment of the present invention for achieving the above object, if the closing price information by date of each individual item is input, calculating the daily item rate of return of each individual item Step 1; A second step of calculating a daily investment ratio of the client when the daily stock portfolio information is received from the client; A third step of extracting a plurality of bootstrap samples using daily stock yield data of each individual stock for a certain period of time; Calculating a mean for each bootstrap sample, calculating an average and a variance of the mean values of the plurality of bootstrap samples, and calculating an average yield and a yield variance of each individual item; And a fifth step of calculating an expected rate of return and an expected risk using the average rate of return and rate of return of each individual item and the daily investment rate of the client.

In addition, the expected return and expected risk financial information service method according to the present invention, if the closing price information by date of each individual item, the closing price information by date of each industry and the closing price information by date of each market, A first step of calculating daily stock returns, daily stock returns for each sector and daily stock returns for each market; A second step of calculating a daily investment ratio of the client when the daily stock portfolio information is received from the client; A third step of applying the daily industry rate of return and the daily market rate of return to a multiple regression equation to predict the predicted rate of return and variance of each individual item; And a fourth step of calculating an expected return rate and an expected risk of the client using the predicted rate of return and variance of each individual item and the daily investment rate of the client.

According to the present invention, there is provided a computer-readable recording medium having recorded thereon a program for executing the above-described expected yield and expected risk financial information service method on a computer.

Hereinafter, an expected return rate and an expected risk financial information service method according to an embodiment of the present invention will be described with reference to the accompanying drawings.

1 is a block diagram illustrating an expected return rate and an expected risk financial information service system to which the present invention is applied.

The service system according to the present invention includes a web server 101, an expected return / expected risk calculation server 102, and a database management system 103. The database management system 103 manages the item database 104, the industry database 105, the ledger database 106, the member database 107, and the yield / risk database 108.

The expected return / expected risk calculation server 102 is connected to the Korea Securities Computerization System 112 through the network 111. From the Korea Securities Computerization System 112, information on each individual item and each industry and market The stock market information is input, and the stock database 104 and the industry database 105 are constructed. In addition, on the basis of the closing price of the date, the sector index by industry and the market index of the market, the yield of the individual stock, the yield of the industry to which the individual stock belongs, and the return of the market to which the individual stock belongs, are calculated in the stock database 104. Save it. The expected return / expected risk calculation server 102 may receive inputs of individual stock quotes and quotes for each industry and market directly from the manager, not the Korea Securities Computer System.

Stock database 104 stores the name of each individual item and the closing price by date and the type of business and market to which the individual item belongs, the stock return of the corresponding item and the industry return of the industry to which the item belongs and the corresponding item Store the market rate of return for the market to which it belongs.

date event Closing price Sectors market Stock Return (rr) Industry Return (irr) Market rate of return (mrr) 2003/08/07 Samsung 407,000 Electric and electronic exchange -0.72% -0.45% 0.27% 2003/08/06 Samsung 410,000 Electric and electronic exchange 2.44% 0.64% 1.30%

The industry database 105 stores the stock index of each industry and the stock index of the market by date as shown in Table 2.

date Sectors Species index Market index 2003/08/07 Electric and electronic 4197.29 709.81 2003/08/06 Electric and electronic 4216.39 707.88

Here, the item return of Table 1 is calculated by applying to Equation 1 below.

Where rr _t is the return on the item at the date t, C _t is the closing price of the date t, and C _t-1 is the closing price of the date t-1. In other words, the return on an item (rr _t ) on any date can be obtained as the ratio of the share price fluctuation of the individual item on that date to the closing price.

Similarly, the industry return on any date t (irr _t ) is obtained as the ratio of the industry's index fluctuations to the industry's index on that date, and the market return on any date t (mrr _t ) and the market's index fluctuations on the date and market Obtained by the ratio of the species index.

The web server 101 provides an expected return rate and expected risk information service to a plurality of clients through the Internet 121. The client is divided into an individual client 122, a securities company client 123, and an enterprise client 124. . Each client provides the portfolio information of its own stock to the web server 101, and the web server 101 calculates and provides the expected return and expected risk of the portfolio to the corresponding client.

Information about each client is stored in the member database 107, and information about the portfolio of shares held by each client is stored in the ledger database 106. The ledger database 106 stores date, customer ID, stock item and stock number information, closing price information for each individual item referring to the item database, investment amount and investment ratio information as shown in Table 3.

date Customer ID event Shares Closing price Investment amount Investment ratio 2003/08/07 A Samsung 60 410,000 16,400,000 0.775 2003/08/07 A Hynix 400 9,350 3,740,000 0.177 2003/08/07 A Chohung Bank 300 3,410 1,023,000 0.048

When the information on the stock held in the ledger database 106 is stored, the expected return / expected risk calculation server 102 calculates the expected return and expected risk of the investor using the information stored in the ledger database 106.

Value at Risk (VAR) is used to calculate expected returns and expected risks. VAR is an estimate of the maximum amount of loss that can occur over a given period of time at a given level of confidence, assuming normal market conditions. In other words, the VAR is the estimated value of the maximum loss in the portfolio held over a period of time, if the market moves in an adverse direction, using a statistical method over a given confidence interval.

For example, if the daily VAR of a portfolio that a financial institution currently builds is 10 billion won at a 99% confidence level, the financial institution may hold more than 10 billion won in the next day by holding this portfolio under normal market conditions. Means 1%.

2 is a flowchart illustrating a process of calculating an expected return and an expected risk of an investor according to the present invention.

First, the daily stock price information and the client's stock portfolio information is received (S10), and as described above, using the received daily stock price information to calculate the daily rate of return, industry rate and market rate of return in the item database and industry database And, the investment ratio of the daily client is calculated and stored in the ledger database (S20).

When the bootstrap calculation is requested from the client (S30), the expected return / expected risk calculation server 102 calculates the expected return and the expected risk by the bootstrap method (S40), and when the multiple regression calculation is requested from the client ( S50), the expected return / expected risk calculation server 102 calculates the expected return and the expected risk by a multiple regression analysis method (S60). In addition, the expected return and the estimated risk calculated by each method are stored in the yield / risk database and provided to the corresponding client (S70).

The bootstrap method uses the historical data to create a large number of bootstrap samples to predict short-term expected returns and expected risks tomorrow or two or three days later, and the multiple regression method uses several different variables (industry yield and market). Rate of return) is used to calculate predictable expected return and predicted risk.

First, a method of calculating an expected return rate and an expected risk using a bootstrap method will be described with reference to FIG. 3.

B bootstrap samples are extracted by using the data of individual stocks accumulated for a certain period of time (S41). In other words, the individual stock return data accumulated over a period of time is X = {X ₁ , X ₂ ,. , X _n } (where n is the number of individual stock returns), then the randomized random sample extracted from the bootstrap sample is X ^* = {X ₁ ^* , X ₂ ^* ,. , X _n ^* }, individual stock returns X = {X ₁ , X ₂ ,. B bootstrap samples (X ^{* 1} , X ^{* 2} , ..., X ^{* b} ) are extracted by repeating B times independently from, _Xn }.

(b = 1, 2,…, B)을 계산하고(S42), 각 개별종목의 평균수익률(

)과 수익률의 분산(

)을 아래의 수학식 2 및 수학식 3과 같이 구한다(S43).And, for each bootstrap sample (X ^{* 1} , X ^{* 2} ,…, X ^{* b} )

(b = 1, 2,…, B) (S42), and the average rate of return for each individual item (

) And variance of return (

) Is obtained as in Equation 2 and Equation 3 below (S43).

As described above, when the average return and the variance of the individual returns are calculated, the expected returns and the expected risks of each client are calculated by using a value at risk (VAR) equation such as Equation 4 and Equation 5.

는 평균수익 률 행렬이다. 수학식 4를 살펴보면, 각 개별종목에 해당하는 투자비율과 붓스트랩 방법을 통한 각 종목의 평균수익률의 곱의 합이 해당 클라이언트의 기대수익률임을 알 수 있다.Where P is the expected return of the client, W is the investment rate matrix for each item according to the client's portfolio, W ^T is the transpose matrix of the investment rate for each item,

Is the average return matrix. Looking at Equation 4, it can be seen that the sum of the investment ratio corresponding to each individual item and the average return of each item through the bootstrap method is the expected return of the client.

Where VAR (P) is the expected risk of the client, Z _α is the confidence level, W is the investment rate matrix for each item according to the client's portfolio, W ^T is the transpose matrix of the investment rate for each item, and Cov (Y) is the variance of the yield. to be. Referring to Equation 5, multiplying the return variance of each individual item by the investment ratio and confidence level of the respective item can calculate the expected risk of the client.

The expected risk and expected return rate of the daily client, calculated by bootstrap method, are stored in the yield / risk database as shown in Table 4 and provided to the corresponding client.

date Customer ID Expected risk Expected Return 2003/08/07 A 0.053 0.038

Next, a method of calculating an expected return rate and an expected risk using a multiple regression analysis method will be described with reference to FIG. 4.

The Capital Asset Pricing Model (CAPM) has been widely used as a model for representing stock movement. This model has the premise that the stock price movement of individual stocks reflects the overall market movement. In other words, the movement of the entire market or the sector to which an individual stock belongs helps to predict the main movement of the individual stock. When modeling this, it is expressed as a multiple linear regression equation as shown in Equation 6 below.

Where rr _jt is the return of the individual item (the share price return on date t for item j), irr _jt is the industry return to which the individual item belongs, mrr _jt is the market return to which the item belongs, irr _jt * mrr _jt Is the entanglement effect of industry returns and market returns, and ε _jt is the error.

A simple expression of the multiple linear regression equation of Equation 6 can be expressed as Equation 7.

Here, Y is the stock return and X is a matrix including various variables that correlate with stock returns, and three variables are used: industry returns, market returns, and industry returns *. Β is a regression coefficient vector representing a linear relationship between Y and X, and ε is an error.

In other words, when calculating the multiple regression analysis, the stock returns are first predicted using the industry returns and the market returns (S61). In other words, β and ε in Equation 7 are calculated by using a least-squares method that finds the least error between the actual observation and the predicted value of the individual stock returns, and predicts the X and Y values.

In addition, the variance covariance matrix (Cov (Y)) of the individual stock returns is calculated by applying tomorrow's forecast x ₀ value (sector yield, market returns, sector returns * market returns) to Equation 8 below (S62).

는 종목수익률의 분산값으로써, 다중회귀분석 방식에 의해 계산된 예측 종목수익률을 이용하여 계산한다.At this time,

Is the variance of stock returns and is calculated using the predicted stock returns calculated by the multiple regression analysis.

When the individual item return prediction value and the variance covariance matrix are calculated, the expected return P is calculated by applying it to Equation 9, and the expected risk VAR (P) is calculated by applying it to Equation 10 (S63).

Here, Cov (Y) is a distributed covariance matrix obtained by Equation 8, W is an investment ratio for each item, and Z _α is a confidence level.

The expected return and the estimated risk calculated by the multiple regression analysis are stored in the yield / risk database and provided to each client.

Expected Return and Expected Risk As described above, the financial information service method may calculate and provide the expected return and risk according to the stock portfolio through the Internet, or may be installed on an individual computer offline to provide the same service to the user. It may be.

Although the technical spirit of the present invention has been described above with reference to the accompanying drawings, it is intended to exemplarily describe the best embodiment of the present invention, but not to limit the present invention. In addition, it is obvious that any person skilled in the art may make various modifications and imitations without departing from the scope of the technical idea of the present invention.

As described above, according to the present invention, the expected return rate and the expected risk of the future are provided to the client by using the return data of each item in the past period, the industry return rate, and the market return data, so that the securities related website is provided to the user. The service quality is improved.

Claims (9)

The computer receives the daily closing price information of each individual item, the closing price information of each industry, the closing price information of each market, and the daily stock portfolio information of the client, and calculates the expected return and expected risk according to the daily stock portfolio. In the method of serving the client, When the closing price information by date of each individual item, the closing price information by date of each industry, and the closing price information by date of each market are inputted, the daily stock returns of the respective stocks, the daily stock returns of the respective stocks and the daily market returns of the respective markets. Calculating a first step; A second step of calculating a daily investment ratio of the client when the daily stock portfolio information is received from the client; A third step of obtaining a multiple linear regression equation using the daily industry rate of return, daily market rate of return and daily item rate of return; A fourth step of predicting an industry return, a market return, and an item return using the multiple linear regression equation; A fifth step of calculating a distributed covariance matrix of the stock returns using the forecast stock returns, the forecast market returns, and the forecast stock returns; A sixth step of calculating an expected return rate and an expected risk of the client using the predicted return rate, the distributed covariance matrix, and the daily investment rate of each individual item constituting the daily stock portfolio; A seventh step of extracting a plurality of bootstrap samples using daily stock returns for each individual stock over a period of time; Calculating an average for each bootstrap sample, calculating an average and a variance of the average values of the plurality of bootstrap samples, and calculating an average yield and a yield variance of each individual item; Expected rate of return and expected risk financial information comprising the ninth step of calculating the expected rate of return and expected risk using the average rate of return and rate of return of each individual item obtained in the eighth step and the daily investment rate of the client Service method. The method of claim 1, wherein the sixth step is Expected return and expected risk financial information service method characterized by calculating the expected return rate of the client by multiplying the predicted rate of return of each individual item and the investment ratio of the respective individual item of the client and then sums for all items. The method of claim 1, wherein the sixth step is Expected return and expected risk financial information service method characterized by calculating the expected risk of the client by multiplying the variance predicted value of each individual item by the investment ratio and the confidence level of the corresponding individual item of the client. The method of claim 1, And a tenth step of providing the client with the expected return and expected risk information calculated in the sixth and ninth steps. The method of claim 1, wherein the ninth step is Expected return and expected risk financial information service method characterized by calculating the expected return rate of the client by multiplying the average rate of return of each individual item and the investment rate of the respective individual item of the client and then adding up for all items. The method of claim 5, wherein the ninth step, Expected return and expected risk financial information service method characterized by calculating the expected risk of the client by multiplying the rate of return of each individual item, the investment ratio and the confidence level of the corresponding individual item of the client and then adding up for all items. On your computer, When the closing price information by date of each individual item, the closing price information by date of each industry, and the closing price information by date of each market are inputted, the daily stock returns of the respective stocks, the daily stock returns of the respective stocks and the daily market returns of the respective markets. Calculating a first step; A second step of calculating a daily investment ratio of the client when the daily stock portfolio information is received from the client; A third step of obtaining a multiple linear regression equation using the daily industry rate of return, daily market rate of return and daily item rate of return; A fourth step of predicting an industry return, a market return, and an item return using the multiple linear regression equation; A fifth step of calculating a distributed covariance matrix of the stock returns using the forecast stock returns, the forecast market returns, and the forecast stock returns; A sixth step of calculating an expected return rate and an expected risk of the client using the predicted return rate, the distributed covariance matrix, and the daily investment rate of each individual item constituting the daily stock portfolio; A seventh step of extracting a plurality of bootstrap samples using daily stock returns data of each individual stock for a certain period of time; Calculating an average for each bootstrap sample, calculating an average and a variance of the average values of the plurality of bootstrap samples, and calculating an average yield and a yield variance of each individual item; Execution of the expected return and expected risk financial information service method including the ninth step of calculating the expected return and the expected risk using the average return and return variance of each individual item obtained in the eighth step and the daily investment ratio of the client A computer-readable recording medium that contains a program for making a program. delete delete

Images