Fund real-time valuation method based on ARIMA + OLS

Abstract

The invention discloses a fund real-time valuation method based on ARIMA + OLS, and the method achieves the efficient and real-time valuation of a fund net value through integrating the time sequence prediction capability of an ARIMA model and the parameter optimization capability of the OLS. The method provided by the invention not only overcomes the limitation of a traditional valuation method, but also improves the precision and timeliness of valuation, and provides a more reliable investment decision basis for investors.

Description

Technical Field

[0001] The present invention relates to the field of fund investment, and in particular to a real-time fund valuation method based on ARIMA+OLS. Background Art

[0002] With the rapid development of the financial market, fund investment has become one of the important choices for investors. However, the accuracy and real-time nature of fund valuation are crucial for investors' decision-making. Traditional fund valuation methods often rely on regularly published net asset value data, which has problems of time lag and insufficient accuracy. In order to solve these problems, the industry urgently needs a more advanced and real-time fund valuation technology.

[0003] As a classic time series analysis method, the ARIMA (autoregressive differential moving average) model has been widely used in the prediction of financial data. It can effectively capture the trend and cyclical characteristics of time series data. However, a single ARIMA model may face the problem of insufficient model adaptability when processing complex financial data. Therefore, combining the least squares method (OLS) for parameter optimization and model calibration can significantly improve the accuracy and stability of valuation.

[0004] At present, most of the existing technical solutions are based on the position method for valuation. The core of this method is to use the most recently disclosed position information and some algorithms to extrapolate the underlying holdings of the fund and their corresponding weights to the calculation date.

[0005] Problems of existing technologies: Poor timeliness: Quarterly and annual fund reports usually have a certain lag and are released less frequently (once a quarter or once a year), making it difficult to reflect the latest changes in the fund's investment portfolio in a timely manner. Therefore, real-time valuations based on these reports may not accurately reflect the fund's current net value.

[0006] Incomplete disclosure of holdings: Quarterly and annual reports usually only disclose the top ten holdings of a fund, which means that investors can only understand a portion of the holdings in the fund's portfolio. At the same time, some important holding details (such as holding costs, holding changes, etc.) may not be disclosed, making it difficult to accurately construct a valuation model. In addition, the fund may also hold some non-publicly traded assets (such as private equity, real estate, etc.), and the value of these assets is even more difficult to accurately estimate. Summary of the invention

[0007] In view of the above problems, the present invention is proposed to provide a real-time fund valuation method based on ARIMA+OLS that overcomes the above problems or at least partially solves the above problems.

[0008] According to one aspect of the present invention, a real-time fund valuation method based on ARIMA+OLS is provided, and the real-time fund valuation method comprises:

[0009] Step S1: Collect historical adjusted net value data of funds and daily index data;

[0010] Step S2: convert the adjusted net value and the adjusted closing price of the index into yields and align them by date;

[0011] Step S3: Calculate the most relevant index corresponding to different funds;

[0012] Step S4: construct the ARMIA model through the fund's adjusted net value time series, and obtain the parameters of the lag term p and the moving average term q by combining the minimum AIC principle;

[0013] Step S5: Construct a data set based on the p, q parameters and market factor data obtained in step S4, and divide it into a training set and a validation set:

[0014] Step S6: Based on the data set obtained in step S5, construct an OLS linear regression model, fit the model, obtain the weights corresponding to each x, and record them with special symbols as model parameters;

[0015] Step S7: Calculate the estimated real-time rate of return on day t+1 according to the model parameters and real-time market factors obtained in step S6;

[0016] Step S8: based on the net value of the fund to be evaluated on day t and the real-time rate of return of the fund to be evaluated on day t+1 estimated in step S7;

[0017] Step S9: Backtest the deviation between the actual value and the estimated value of the fund.

[0018] Optionally, the step S1: collecting the fund's historical adjusted net asset value data and index daily line data specifically includes: collecting the fund's historical adjusted unit net asset value data and index daily line data with intraday trading data.

[0019] Optionally, the step S2: converting the adjusted net value and the adjusted closing price of the index into a yield and aligning them by date specifically includes:

[0020] For historical data, the fund's adjusted net value and the real-time index closing price with intraday trading are aligned according to the transaction date dimension and converted into yield.

[0021] Optionally, the step S3: calculating the most relevant index corresponding to different funds specifically includes: calculating the index with the highest correlation coefficient for different funds according to the correlation coefficient matrix.

[0022] Optionally, the dividing into a training set and a validation set specifically includes: dividing the training set and the validation set according to an 8:2 ratio.

[0023] Optionally, the step S7: calculating the estimated real-time rate of return on day t+1 according to the model parameters and real-time market factors obtained in step S6 specifically includes:

[0024] The real-time rate of return calculation formula is as follows:

[0025]

[0026] The first part is the dot product of the weight of the lag term and the rate of return, the second part is the dot product of the weight of the moving average term and the rate of return, the third part is the dot product of the weight of the market factor and the rate of return, and b is a constant term.

[0027] Optionally, the step S8: based on the net value of the fund to be evaluated on day t and the real-time yield of the fund to be evaluated on day t+1 estimated in step S7 specifically includes:

[0028] Estimate the real-time net value of the fund to be valued on day t+1;

[0029] The real-time net worth calculation formula is as follows:

[0030]

[0031] Among them, y t is the net value of the fund to be valued on day t, It is the estimated real-time rate of return of the fund to be valued on day t+1.

[0032] Optionally, the step S9: backtesting the deviation between the actual fund value and the estimated value specifically includes: backtesting the deviation between the estimated fund net value and the actual net value to evaluate whether the online conditions are met.

[0033] The present invention provides a fund real-time valuation method based on ARIMA+OLS, and the fund real-time valuation method comprises: step S1: collecting historical adjusted net value data of funds and index daily line data; step S2: converting the adjusted net value and the adjusted closing price of the index into yields and aligning them by date; step S3: calculating the most relevant index corresponding to different funds; step S4: constructing an ARMIA model through the fund adjusted net value time series, and combining the minimum AIC principle to obtain the lag term p and the moving average term q parameters; step S5: based on the p and q parameters obtained in step S4, and market factor data to construct a data set, and divide it into a training set and a validation set: Step S6: Based on the data set obtained in Step S5, construct an OLS linear regression model, fit the model, obtain the weights corresponding to each x, and record them with special symbols as model parameters; Step S7: Calculate the estimated real-time yield on the t+1th day according to the model parameters and real-time market factors obtained in Step S6; Step S8: Based on the net value of the fund to be valued on the tth day and the real-time yield of the fund to be valued on the t+1th day estimated in Step S7; Step S9: Backtest the deviation between the actual value of the fund and the estimated value. By integrating the time series prediction ability of the ARIMA model and the parameter optimization ability of OLS, efficient and real-time valuation of the net value of the fund is achieved. This method not only overcomes the limitations of traditional valuation methods, but also improves the accuracy and timeliness of valuation, providing investors with a more reliable basis for investment decisions.

[0034] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0036] Figure 1 A flowchart of a real-time fund valuation method based on ARIMA+OLS is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0038] The terms "comprises" and "having" and any variations thereof in the description embodiments, claims and drawings of the present invention are intended to cover non-exclusive inclusions, for example, including a series of steps or units.

[0039] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0040] A real-time fund valuation method based on ARIMA+OLS includes:

[0041] Step S1: Collect historical adjusted net value data of funds and daily index data

[0042] Step S2: Convert the adjusted net value and index adjusted closing price into yield and align them by date

[0043] Step S3: Calculate the most relevant index corresponding to different funds.

[0044] Step S4: Construct the ARMIA model through the fund's adjusted net asset value time series, and combine the minimum AIC principle to obtain the p (lag term) and q (moving average term) parameters.

[0045] Step S5: Construct a data set based on the p, q parameters and market factor data obtained in step S4, and split the training set and validation set into 8:2:

[0046] Step S6: Based on the data set obtained in step S5, construct an OLS linear regression model, fit the model, obtain the weights corresponding to each x, and record them with special symbols as model parameters.

[0047] Step S7: Calculate the estimated real-time yield on day t+1 according to the model parameters and real-time market factors obtained in step S6. The real-time yield calculation formula is as follows:

[0048]

[0049] The first part is the dot product of the weight of the lag term and the rate of return, the second part is the dot product of the weight of the moving average term and the rate of return, the third part is the dot product of the weight of the market factor and the rate of return, and b is a constant term.

[0050] Step S8: Estimate the real-time net value of the fund to be evaluated on day t+1 based on the net value of the fund to be evaluated on day t and the real-time rate of return of the fund to be evaluated on day t+1 estimated in step S7. The real-time net value calculation formula is as follows:

[0051]

[0052] Among them, y t is the net value of the fund to be valued on day t, It is the estimated real-time rate of return of the fund to be valued on day t+1.

[0053] Step S9: Backtest the deviation between the actual value and the estimated value of the fund.

[0054] A real-time fund valuation method based on ARIMA+OLS is characterized by implementing the above steps with an example as follows.

[0055] A real-time fund valuation method based on ARIMA+OLS, characterized in that, in step S1, historical adjusted unit net value data of the fund and index daily line data with intraday trading data are collected;

[0056] The sample table of fund historical adjusted unit net value data is as follows:

[0057]

[0058]

[0059] The sample table of industry index data with intraday trading is as follows:

[0060]

[0061] A real-time fund valuation method based on ARIMA+OLS, characterized in that, in step S2, for historical data, the adjusted net value of the fund and the closing price of the real-time index with intraday transactions need to be aligned according to the transaction date dimension and converted into a rate of return;

[0062]

[0063]

[0064] A real-time fund valuation method based on ARIMA+OLS, characterized in that, in step S3, the index with the highest correlation coefficient is calculated for different funds respectively through the correlation coefficient matrix;

[0065] Through correlation coefficient calculation, we found that the correlation coefficient of "China Merchants Industry Selected Stock Fund" is the highest among the CSI 300, Shanghai Composite Index, CSI 500 and Dividend Index, with the Dividend Index being 92%.

[0066] A real-time fund valuation method based on ARIMA+OLS is characterized in that, in step S4, an ARIMA model is constructed based on the fund weighted net value yield time series obtained in S2, and p and q parameters are obtained according to the AIC principle:

[0067]

[0068] Based on the ARIMA model and the AIC minimum principle, the p and q parameters of the fund are calculated to be (2, 2).

[0069] Special note: p is the lag term, which refers to the data lagged 2 orders at the current time, and q is the moving average term, which is the average of the lagged 2 order data. For example, for the trading day 20240710, when p=2, the value is the net asset value rate of return corresponding to the 20240708 fund, which is 0.002552. When q=2, the value is the average of the lagged 1 order and lagged 2 order data, which is the average of the net asset value rate of return of 20240708 and 20240709: (0.002552+0.002932) / 2=0.002742.

[0070] In step S5, a data set is constructed based on the p and q parameters and the market factor data obtained from step S4, and the training set and the validation set are divided into 8:2.

[0071] Based on the data set obtained from S5 in step S6, an OLS linear regression model is constructed, the model is fitted, and the weights corresponding to each x are obtained and recorded;

[0072] For example, the final weight parameters obtained according to the OLS model are {"p1":0.00332, "q1":0.00002..}

[0073] A real-time fund valuation method based on ARIMA+OLS, characterized in that the real-time yield of the fund to be valued on the t+1 day is estimated according to the dynamic weights of each asset on the t day calculated in step S7.

[0074] For example, to value the fund at the trading time of 204-07-11 10:00:00, it is necessary to find the weighted rate of return and the weight parameter respectively, and then combine the rate of return of the dividend index at 204-07-11 10:00:00 to estimate the rate of return:

[0075] r`=0.00332*0.002932+0.00002*0.002932+...

[0076] A real-time fund valuation method based on ARIMA+OLS, characterized in that the real-time estimated net value is calculated based on the net value of the fund to be valued on the tth day and the rate of return estimated in step S8.

[0077] For example, if the net value of the previous day is 1.2503 and the estimated rate of return r` is 1.52%, then the real-time estimated net value is: 1.25*(1+1.52%)=1.2693.

[0078] In step S9, the deviation between the estimated fund net value and the actual net value is backtested to evaluate whether the upper limit conditions are met.

[0079] The present invention comprehensively considers the correlation between the current fund itself and each index, and proposes a new fund real-time valuation method in combination with the ARIMA model and the OLS model. The method first calculates the market factors most relevant to the fund based on the historically published net value of the fund and the net value series of each market index, and then calculates the p, q values ​​(i.e., (lag term and moving average term)) based on the ARIMA model through the minimum AIC principle, and then uses the OLS model to take the target fund asset yield as the y value, the own lag term, the own moving average term and the yield of the latest market factor as x, construct a regression model, segment the data set, and fit the model parameters of each x.

[0080] Based on the model parameters of each x, during the intraday trading stage, the fund yield is estimated by combining the yield of the real-time intraday trading market factors to obtain the estimated net value of the fund.

[0081] Beneficial effects: The present invention proposes a real-time fund valuation method based on ARIMA+OLS. By integrating the time series prediction capability of the ARIMA model and the parameter optimization capability of OLS, efficient and real-time valuation of the fund net value is achieved. The method of the present invention not only overcomes the limitations of traditional valuation methods, but also improves the accuracy and timeliness of valuation, providing investors with a more reliable basis for investment decision-making.

[0082] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A real-time fund valuation method based on ARIMA+OLS, characterized in that: The fund real-time valuation method includes: Step S1: Collect historical adjusted net value data of funds and daily index data; Step S2: convert the adjusted net value and the adjusted closing price of the index into yields and align them by date; Step S3: Calculate the most relevant index corresponding to different funds; Step S4: construct the ARMIA model through the fund's adjusted net value time series, and obtain the parameters of the lag term p and the moving average term q by combining the minimum AIC principle; Step S5: Construct a data set based on the p, q parameters and market factor data obtained in step S4, and divide it into a training set and a validation set: Step S6: Based on the data set obtained in step S5, construct an OLS linear regression model, fit the model, obtain the weights corresponding to each x, and record them with special symbols as model parameters; Step S7: Calculate the estimated real-time rate of return on day t+1 according to the model parameters and real-time market factors obtained in step S6; Step S8: based on the net value of the fund to be evaluated on day t and the real-time rate of return of the fund to be evaluated on day t+1 estimated in step S7; Step S9: Backtest the deviation between the actual value and the estimated value of the fund.

2. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The step S1: collecting the fund's historical adjusted net value data and index daily line data specifically includes: Collect the fund's historical adjusted unit net asset value data and index daily data with intraday trading data.

3. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The step S2: converting the adjusted net value and the adjusted closing price of the index into a yield and aligning them by date specifically includes: For historical data, the fund's adjusted net value and the real-time index closing price with intraday trading are aligned according to the transaction date dimension and converted into yield.

4. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The step S3: calculating the most relevant index corresponding to different funds specifically includes: calculating the index with the highest correlation coefficient for different funds according to the correlation coefficient matrix.

5. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The dividing into a training set and a validation set specifically includes: dividing the training set and the validation set according to 8:

2.

6. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The step S7: calculating the estimated real-time yield on the t+1 day according to the model parameters and real-time market factors obtained in step S6 specifically includes: The real-time rate of return calculation formula is as follows: The first part is the dot product of the weight of the lag term and the rate of return, the second part is the dot product of the weight of the moving average term and the rate of return, the third part is the dot product of the weight of the market factor and the rate of return, and b is a constant term.

7. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The step S8: based on the net value of the fund to be evaluated on day t and the real-time yield of the fund to be evaluated on day t+1 estimated in step S7 specifically includes: Estimate the real-time net value of the fund to be valued on day t+1; The real-time net worth calculation formula is as follows: Among them, y t is the real net value of the fund to be valued on day t, It is the estimated real-time rate of return of the fund to be valued on day t+1.

8. The real-time fund valuation method based on ARIMA+OLS according to claim 1, characterized in that: The step S9: backtesting the deviation between the actual fund value and the estimated value specifically includes: backtesting the deviation between the estimated fund net value and the actual net value to evaluate whether the online conditions are met.

Images