Fund benchmark determination method based on historical net value and semantic vector

Abstract

The invention discloses a fund benchmark determination method based on a historical net value and a semantic vector. The benchmark determination method comprises the following steps: collecting fund historical reweighting net value data and index daily data; the reweighting net value data and the index daily line data are converted into return rates, and the return rates are aligned according to dates to serve as historical market vectors for describing funds and indexes; fund investment strategy description and index compilation method brief description data are collected; carrying out vector conversion on fund investment strategy description and exponential compilation method brief description data by utilizing a semantic vector model, and constructing semantic vectors; and calculating a fund evaluation index based on the obtained optimal matching vector. And accurate positioning and dynamic adjustment of the fund benchmark are realized, so that high-standard requirements of investors on fund evaluation are met, and healthy and stable development of the financial market is promoted.

Description

Technical Field

[0001] The present invention relates to the field of financial investment, and in particular to a fund benchmark determination method based on historical net worth and semantic vectors. Background Art

[0002] In the current financial investment field, with the rapid development of big data and artificial intelligence technology, investors have put forward higher requirements for the accuracy and timeliness of fund performance evaluation. Traditionally, the selection of fund benchmarks mostly depends on market indexes or the average performance of similar funds. Although this method is simple, it is often difficult to fully reflect the actual investment strategy and risk-return characteristics of the fund. Especially when the market environment is complex and changeable, and a single market index is difficult to accurately capture the multi-dimensional performance of the fund, it is particularly important to seek a more scientific and personalized fund benchmark.

[0003] In the existing technical solutions, the methods for finding fund benchmarks are mainly concentrated in two categories. One is the traditional method based on market indexes or the average performance of similar funds. This method selects market indexes close to the fund's investment scope or calculates the average performance of similar funds as the benchmark. The other is to use quantitative analysis techniques, such as factor models and cluster analysis, to conduct in-depth mining of the fund's historical data to identify fund groups with similar characteristics, thereby determining the benchmark. Summary of the invention

[0004] In view of the above problems, the present invention is proposed to provide a fund benchmark determination method based on historical net worth and semantic vectors that overcomes the above problems or at least partially solves the above problems.

[0005] According to one aspect of the present invention, a fund benchmark determination method based on historical net worth and semantic vectors is provided, the benchmark determination method comprising:

[0006] Step S1: Collect historical adjusted net value data of funds and daily index data;

[0007] Step S2: convert the adjusted net value data and the index daily line data into yields, and align them by date as the historical market vectors describing the fund and the index respectively;

[0008] Step S3: Collect data on fund investment strategy description and index compilation method;

[0009] Step S4: using the semantic vector model to perform vector conversion on the fund investment strategy description and the index compilation method summary data respectively, and construct a semantic vector;

[0010] Step S5: respectively concatenate the fund and index historical market vectors obtained in step S2 and the fund and index semantic vectors obtained in step S3 as complete vectors describing the fund and index;

[0011] Step S6: using the correlation coefficient to calculate the index vector most relevant to the fund in the index as the final best matching index;

[0012] Step S7: Calculate the fund evaluation index based on the best matching vector obtained in step S6.

[0013] Optionally, the fund evaluation indicators specifically include: excess return, Treynor ratio and Sharpe ratio.

[0014] 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.

[0015] Optionally, the step S2: converting the adjusted net value data and the index daily line data into yields and aligning them by date as historical market vectors describing the fund and the index respectively specifically includes:

[0016] 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 a yield.

[0017] Optionally, the step S3: collecting fund investment strategy description and index compilation method summary data specifically includes: collecting fund investment strategy data, and investment or selection object data of each index.

[0018] Optionally, the step S6: using the correlation coefficient to calculate the index vector most relevant to the fund in the index as the final best matching index specifically includes:

[0019] The complete vectors of all funds and the complete index vectors obtained from step S5 are used to calculate the most relevant index benchmark for each fund.

[0020] The present invention provides a fund benchmark determination method based on historical net value and semantic vector, the benchmark determination method comprises: step S1: collecting historical adjusted net value data of funds and daily index data; step S2: converting the adjusted net value data and daily index data into yields, and aligning them by date as historical market vectors describing funds and indexes; step S3: collecting fund investment strategy descriptions and index compilation method brief data; step S4: using semantic vector models to perform vector conversion on fund investment strategy descriptions and index compilation method brief data, and construct semantic vectors; step S5: splicing the fund and index historical market vectors obtained in step S2 and the fund and index semantic vectors obtained in step S3 as complete vectors describing funds and indexes; step S6: using correlation coefficients to calculate the index vector most relevant to the fund in the index as the final best matching index; step S7: calculating fund evaluation indicators based on the best matching vector obtained in step S6. Accurate positioning and dynamic adjustment of fund benchmarks are achieved to meet investors' high standards for fund evaluation and promote the healthy and stable development of financial markets.

[0021] 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

[0022] 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.

[0023] Figure 1 A flow chart of a method for determining a fund benchmark based on historical net worth and semantic vectors provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] 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.

[0025] 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.

[0026] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0027] like Figure 1 As shown, a method for finding a fund benchmark based on historical net worth and semantic vectors includes:

[0028] Step S1: Collect historical adjusted net value data of funds and daily index data

[0029] Step S2: Convert the adjusted net value and the adjusted closing price of the index into yields and align them by date as the historical market vectors describing the fund and the index respectively.

[0030] Step S3: Collect data on fund investment strategy description and index compilation method

[0031] Step S4: Use the semantic vector model to perform vector conversion on the fund investment strategy description and index compilation method to construct a semantic vector.

[0032] Step S5: respectively concatenate the fund and index historical market vectors obtained in step S2 and the fund and index semantic vectors obtained in step S3 as complete vectors describing the fund and index.

[0033] Step S6: Calculate the index vector most relevant to the fund among many indexes using the correlation coefficient as the final best matching index.

[0034] Step S7: Based on the best matching vector obtained in step S6, derivative applications such as excess return, Treynor ratio, Sharpe ratio and other fund evaluation indicators can be further calculated.

[0035] In step S1, the historical adjusted net asset value data of the fund and the daily index data with intraday trading data are collected;

[0036] The sample table of fund historical adjusted unit net value data is as follows:

[0037]

[0038] The sample table of industry index data with intraday trading is as follows:

[0039]

[0040] 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.

[0041]

[0042] In step S3, the investment strategy data of the fund and the investment or selection object data of each index are collected.

[0043]

[0044]

[0045]

[0046] In step S4, the investment object obtained in step S3 is converted into a vector using the semantic vector.

[0047] The following is an example of data obtained after text vector conversion.

[0048]

[0049] Based on the yield vector obtained from step S2 and the semantic vector obtained from step S4 in step S5, the vectors are concatenated;

[0050] For example, for the China Merchants Industry Select Stock Fund, its daily rate of return vector is: [0.002101, 0.001818, -0.007670, ...]

[0051] Its semantic vector is: [0.3259, 0.5523, 0.62311...], and the vector that comprehensively describes the fund is [0.002101, 0.001818, -0.007670, ..., 0.3259, 0.5523, 0.62311...].

[0052] According to the method of the present invention, a complete vector similar to other funds and other indexes is obtained.

[0053] Based on the complete vectors of all funds and the complete index vectors obtained from step S5 in step S6, the most relevant index benchmarks for each fund are calculated respectively.

[0054] For example, for China Merchants Industry Select Stock Fund, after calculation, the most relevant index is the dividend index, and the correlation coefficient is 92.34%.

[0055] Based on the benchmark vector found in step S7, the fund's indicators such as excess return, Sharpe ratio, Treynor ratio, etc. can be derived and calculated, and then some derivative applications can be performed.

[0056] Beneficial effects: The present invention proposes a new method for finding fund benchmarks based on historical net value and semantic vectors, which aims to comprehensively and deeply explore the historical performance and intrinsic characteristics of funds by integrating financial time series analysis and natural language processing technology, and realize accurate positioning and dynamic adjustment of fund benchmarks, so as to meet investors' high standards for fund evaluation and promote the healthy and stable development of the financial market.

[0057] 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 fund benchmark determination method based on historical net worth and semantic vector, characterized in that: The benchmark determination method comprises: Step S1: Collect historical adjusted net value data of funds and daily index data; Step S2: convert the adjusted net value data and the index daily line data into yields, and align them by date as the historical market vectors describing the fund and the index respectively; Step S3: Collect data on fund investment strategy description and index compilation method; Step S4: using the semantic vector model to perform vector conversion on the fund investment strategy description and the index compilation method summary data respectively, and construct a semantic vector; Step S5: respectively concatenate the fund and index historical market vectors obtained in step S2 and the fund and index semantic vectors obtained in step S3 as complete vectors describing the fund and index; Step S6: using the correlation coefficient to calculate the index vector most relevant to the fund in the index as the final best matching index; Step S7: Calculate the fund evaluation index based on the best matching vector obtained in step S6.

2. A fund benchmark determination method based on historical net worth and semantic vector according to claim 1, characterized in that: The fund evaluation indicators specifically include: excess return, Treynor ratio and Sharpe ratio.

3. A fund benchmark determination method based on historical net worth and semantic vector according to claim 1, characterized in that: 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.

4. A method for determining a fund benchmark based on historical net worth and semantic vectors according to claim 1, characterized in that: The step S2: converting the adjusted net value data and the index daily line data into yields and aligning them by date as the historical market vectors describing the fund and the index respectively, specifically includes: 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 a yield.

5. The method for determining a fund benchmark based on historical net worth and semantic vector according to claim 1, characterized in that: The step S3: collecting fund investment strategy description and index compilation method summary data specifically includes: collecting fund investment strategy data and investment or selection object data of each index.

6. A method for determining a fund benchmark based on historical net worth and semantic vectors according to claim 1, characterized in that: The step S6: using the correlation coefficient to calculate the index vector most relevant to the fund in the index as the final best matching index specifically includes: The complete vectors of all funds and the complete index vectors obtained from step S5 are used to calculate the most relevant index benchmark for each fund.

Images