Investor emotion decoupling method and system based on deviation analysis

Abstract

The invention relates to an investor emotion decoupling method and system based on deviation analysis, and the method comprises the steps: collecting investor emotion data of an investor in a preset time period, the investor emotion data comprising an original individual share emotion value, an original industry emotion value and an original market emotion value; performing emotion deviation analysis on the original industry emotion value according to the original market emotion value to obtain a pure industry emotion value after market emotion influence is eliminated; performing emotion deviation analysis on the original individual share emotion value according to the original market emotion value to obtain a first individual share emotion value after market emotion influence is eliminated; performing emotion deviation analysis on the first individual share emotion value according to the pure industry emotion value to obtain a pure individual share emotion value after market emotion and industry emotion influences are eliminated, and eliminating the influence of market emotion on industry emotion and the influence of market emotion and industry emotion on individual share emotion through an emotion deviation analysis technology to obtain a pure individual share emotion value; therefore, the data accuracy of the investor emotion is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an investor sentiment decoupling method and system based on deviation analysis. Background Art

[0002] Investor sentiment is the part of the asset market that deviates from the expected average. The part above the average is called bullish expectations, and the part below the average can be called bearish expectations. Investors may not know exactly what the expected average value in the market is, but this expectation exists objectively.

[0003] At present, the measurement methods of investor sentiment mainly include sentiment index method and public opinion analysis method. The sentiment index method covers a variety of approaches: for example, based on turnover rate, a sentiment index represented by trading frequency is established; through principal component analysis, the first principal component of multiple alternative indicators is used to measure investor sentiment, although this may contain errors that are irrelevant to future stock returns; individual investors have a herd phenomenon, and market liquidity can be indirectly reflected by stock trading volume. In addition, using sentiment proxy variables such as closed-end fund discount rate, principal component analysis is used to construct an investor sentiment index; the impact of irrational emotions on the stock market is studied through the net inflow of open-end stock funds. The public opinion analysis method uses text mining technology to obtain data from multiple channels such as media reports, company announcements and social networks. After measuring the investor sentiment value, further analysis of the relationship between investor sentiment and financial markets under different conditions is becoming a research hotspot.

[0004] However, the current analysis of the relationship between investor sentiment and financial markets only focuses on sentiment analysis at a certain level and fails to fully consider the mutual influence between emotions at different levels, resulting in inaccurate data on investor sentiment values. Summary of the invention

[0005] The present invention provides an investor sentiment decoupling method and system based on deviation analysis, which eliminates the impact of market sentiment on industry sentiment, as well as the impact of market sentiment and industry sentiment on individual stock sentiment through sentiment deviation analysis technology, thereby improving the data accuracy of investor sentiment.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] First, a method for decoupling investor sentiment based on deviation analysis is provided, including:

[0008] Collect investor sentiment data of investors within a preset time period, including original individual stock sentiment value, original industry sentiment value and original market sentiment value; original market sentiment value reflects investors' investment sentiment towards the entire stock market; original industry sentiment value reflects investors' investment sentiment towards a specific industry in the entire stock market; original individual stock sentiment value reflects investors' investment sentiment towards a specific stock in a specific industry;

[0009] Perform sentiment deviation analysis on the original industry sentiment value based on the original market sentiment value to obtain the pure industry sentiment value after eliminating the influence of market sentiment;

[0010] According to the original market sentiment value, the original stock sentiment value is analyzed for sentiment deviation, and the first stock sentiment value after eliminating the influence of market sentiment is obtained;

[0011] Based on the pure industry sentiment value, the sentiment deviation analysis is performed on the first stock sentiment value to obtain the pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

[0012] Furthermore,

[0013] The original market sentiment value includes the positive sentiment value PM, the neutral sentiment value MM and the negative sentiment value NM; and PM+MM+NM=1;

[0014] The original industry sentiment value includes the positive sentiment value PI k , neutral sentiment value MI k and negative sentiment value NI k , k represents a specific industry; and PI k +MI k +NI k =1;

[0015] The original stock sentiment value includes the positive sentiment value PS m , neutral sentiment value MS m and negative sentiment value NS m , m represents a specific stock in a specific industry k; and PS m +MS m +NS m =1.

[0016] Furthermore, the original industry sentiment value is subjected to sentiment deviation analysis based on the original market sentiment value to obtain the pure industry sentiment value after eliminating the influence of market sentiment, including:

[0017] The market sentiment factor is calculated based on the original market sentiment value;

[0018] Based on the market sentiment factors, the original industry sentiment value is subjected to sentiment deviation analysis to obtain the pure industry sentiment value after eliminating the influence of market sentiment.

[0019] Furthermore, the market sentiment factors are calculated based on the original market sentiment values, including:

[0020] The calculation formula for obtaining the market sentiment factor MF is:

[0021] ;

[0022] Substitute the positive sentiment value PM and negative sentiment value NM of the original market sentiment value into the market sentiment factor MF calculation formula to obtain the market sentiment factor MF.

[0023] Furthermore, the original industry sentiment value is analyzed for sentiment deviation based on the market sentiment factor to obtain the pure industry sentiment value after eliminating the influence of market sentiment, including:

[0024] Set the preset sentiment bias formula as:

[0025] ; Among them, r is MF, the observed value is the original industry sentiment value, and the true value is the pure industry sentiment value after eliminating the influence of market sentiment; the pure industry sentiment value includes the positive sentiment value , neutral sentiment value and negative sentiment ;

[0026] Set the preset emotion conversion rule to only convert neutral emotion to negative emotion or neutral emotion to positive emotion. and The emotional constraint formula is:

[0027] ;

[0028] ;

[0029] Determine whether MF>0 or MF<0;

[0030] If MF>0, determine PI k Greater than , and NI k Less than ; ;

[0031] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0032] ;

[0033] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0034] ;

[0035] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0036] ;

[0037] If MF<0, determine PI k Less than , and NI k Greater than ; ;

[0038] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0039] ;

[0040] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0041] ;

[0042] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0043] .

[0044] Furthermore, the sentiment deviation analysis is performed on the first stock sentiment value according to the pure industry sentiment value to obtain the pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment, including:

[0045] The calculation formula for obtaining the industry sentiment factor IF is:

[0046] ;

[0047] The pure industry sentiment value and Substitute into the market sentiment factor IF calculation formula to obtain the industry sentiment factor IF;

[0048] According to the market sentiment factor IF, the sentiment deviation analysis is performed on the first stock sentiment value to obtain the pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

[0049] Secondly, an investor sentiment decoupling system based on deviation analysis is provided, including:

[0050] The investor sentiment collection module is used to collect investor sentiment data of investors within a preset time period. The investor sentiment data includes original individual stock sentiment values, original industry sentiment values ​​and original market sentiment values. The original market sentiment values ​​reflect investors' investment sentiments towards the entire stock market. The original industry sentiment values ​​reflect investors' investment sentiments towards a specific industry in the entire stock market. The original individual stock sentiment values ​​reflect investors' investment sentiments towards a specific stock in a specific industry.

[0051] The industry sentiment decoupling module is used to perform sentiment deviation analysis on the original industry sentiment value based on the original market sentiment value to obtain the pure industry sentiment value after eliminating the influence of market sentiment;

[0052] The individual stock sentiment decoupling module is used to perform sentiment deviation analysis on the original individual stock sentiment value based on the original market sentiment value to obtain the first stock sentiment value after eliminating the impact of market sentiment; and to perform sentiment deviation analysis on the first stock sentiment value based on the pure industry sentiment value to obtain the pure individual stock sentiment value after eliminating the impact of market sentiment and industry sentiment.

[0053] Furthermore,

[0054] The original market sentiment value includes the positive sentiment value PM, the neutral sentiment value MM and the negative sentiment value NM; and PM+MM+NM=1;

[0055] The original industry sentiment value includes the positive sentiment value PI k , neutral sentiment value MI k and negative sentiment value NI k , k represents a specific industry; and PI k +MI k +NI k =1;

[0056] The original stock sentiment value includes the positive sentiment value PS m , neutral sentiment value MS m and negative sentiment value NS m , m represents a specific stock in a specific industry k; and PS m +MS m +NS m =1.

[0057] Furthermore, the industry sentiment decoupling module includes:

[0058] Market sentiment factor calculation unit, configuration unit, judgment unit and industry sentiment decoupling processing unit;

[0059] The market sentiment factor calculation unit is used to obtain the market sentiment factor MF calculation formula:

[0060] ;

[0061] Substitute the positive sentiment value PM and the negative sentiment value NM of the original market sentiment value into the market sentiment factor MF calculation formula to obtain the market sentiment factor MF;

[0062] Configuration unit, used to set the preset sentiment bias formula is:

[0063] ; Among them, r is MF, the observed value is the original industry sentiment value, and the true value is the pure industry sentiment value after eliminating the influence of market sentiment; the pure industry sentiment value includes the positive sentiment value , neutral sentiment value and negative sentiment ;

[0064] The configuration unit is also used to set the preset emotion conversion rule to only convert the neutral emotion into the negative emotion or the neutral emotion into the positive emotion, and obtain the preset emotion conversion rule. and The emotional constraint formula is:

[0065] ;

[0066] ;

[0067] A judging unit, used for judging whether MF>0 or MF<0;

[0068] Industry sentiment decoupling processing unit, used to determine PI if MF>0 k Greater than , and NI k Less than ; ;

[0069] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0070] ;

[0071] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0072] ;

[0073] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0074] ;

[0075] The industry sentiment decoupling processing unit is also used to determine PI if MF < 0 k Less than , and NI k Greater than ; ;

[0076] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0077] ;

[0078] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0079] ;

[0080] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0081] .

[0082] The beneficial effects achieved by the present invention are:

[0083] Investor sentiment data of investors within a preset time period are collected, and the investor sentiment data include original individual stock sentiment values, original industry sentiment values ​​and original market sentiment values; the original market sentiment values ​​reflect investors' investment sentiment for the entire stock market; the original industry sentiment values ​​reflect investors' investment sentiment for a specific industry in the entire stock market; the original individual stock sentiment values ​​reflect investors' investment sentiment for a specific stock in a specific industry; sentiment deviation analysis is performed on the original industry sentiment values ​​based on the original market sentiment values ​​to obtain pure industry sentiment values ​​after eliminating the influence of market sentiment; sentiment deviation analysis is performed on the original individual stock sentiment values ​​based on the original market sentiment values ​​to obtain the first stock sentiment value after eliminating the influence of market sentiment; sentiment deviation analysis is performed on the first stock sentiment value based on the pure industry sentiment value to obtain the pure individual stock sentiment value after eliminating the influence of market sentiment and industry sentiment. The influence of market sentiment on industry sentiment, as well as the influence of market sentiment and industry sentiment on individual stock sentiment are eliminated through sentiment deviation analysis technology, thereby improving the data accuracy of investor sentiment. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a flow chart of the investor sentiment decoupling method based on deviation analysis of the present invention;

[0085] Figure 2 It is a structural diagram of the investor sentiment decoupling system based on deviation analysis of the present invention;

[0086] Figure 3 This is a structural diagram of the industry emotion decoupling module of the present invention. DETAILED DESCRIPTION

[0087] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0088] like Figure 1 As shown, an embodiment of the present invention provides an investor sentiment decoupling method based on deviation analysis, comprising:

[0089] 101, collecting investor sentiment data of investors within a preset time period, the investor sentiment data including original individual stock sentiment values, original industry sentiment values, and original market sentiment values;

[0090] In this embodiment, by measuring the investment sentiment of investors towards the stock market within a preset time period, the investor sentiment measurement method mainly uses the existing sentiment index method and public opinion analysis method; the sentiment index method can be based on the turnover rate and establish a sentiment index represented by the transaction frequency; through principal component analysis, the first principal component of multiple alternative indicators is used to measure investor sentiment, etc. The public opinion analysis method uses text mining technology to obtain investor sentiment data from multiple channels such as media reports, company announcements and social networks;

[0091] Investor sentiment data includes original individual stock sentiment value, original industry sentiment value and original market sentiment value; original market sentiment value reflects investors' investment sentiment towards the entire stock market; original industry sentiment value reflects investors' investment sentiment towards a specific industry in the entire stock market; original individual stock sentiment value reflects investors' investment sentiment towards a specific stock in a specific industry;

[0092] Specifically, the original market sentiment value includes a positive sentiment value PM, a neutral sentiment value MM, and a negative sentiment value NM; and PM+MM+NM=1;

[0093] The original industry sentiment value includes the positive sentiment value PI k , neutral sentiment value MI k and negative sentiment value NI k , k represents a specific industry; and PI k +MI k +NI k =1;

[0094] The original stock sentiment value includes the positive sentiment value PS m , neutral sentiment value MS m and negative sentiment value NS m , m represents a specific stock in a specific industry k; and PS m +MS m +NS m =1;

[0095] Positive sentiment is associated with rising stock prices because investors' optimism prompts them to increase investment, pushing the stock market upward; negative sentiment is associated with falling stock prices because investors' pessimism leads to increased selling, pushing the stock market downward; neutral sentiment also has its importance in the market. It usually indicates that investors are taking a wait-and-see attitude towards future trends. Understanding neutral sentiment can help identify potential turning points in the market or stable periods of trends.

[0096] 102, performing sentiment deviation analysis on the original industry sentiment value according to the original market sentiment value, and obtaining a pure industry sentiment value after eliminating the influence of market sentiment;

[0097] In this embodiment, the sentiment of investors towards the overall stock market will have a significant impact on the sentiment of each industry; specifically, when the market sentiment is positive, investors are generally optimistic about the entire stock market, and this optimism will be transmitted to each industry, making investors have more positive expectations for the prospects of each industry; on the contrary, when the market sentiment is negative, investors' pessimism about the market will also spread to each industry, causing expectations for the industry to become negative;

[0098] Similarly, investors' sentiment towards a particular industry will further affect their sentiment towards individual stocks in that industry; when investors are optimistic about an industry, they are more likely to be optimistic about the performance of individual stocks in that industry and increase their investment in those stocks; and when they are pessimistic about an industry, investors' confidence in individual stocks in that industry will also be weakened, and they will reduce their investment or choose to sell stocks;

[0099] Therefore, the mutual transmission and influence of investor sentiment at the market, industry and individual stock levels is a progressive process. This multi-level sentiment transmission mechanism has an important impact on investment decisions. In order to accurately capture and analyze the changes in investor sentiment at different levels, it is crucial to understand the mechanism of this sentiment transmission and its impact on investment decisions. For industry sentiment values, the impact of market sentiment needs to be eliminated. The specific process is as follows:

[0100] The calculation formula for obtaining the market sentiment factor MF is:

[0101] ;

[0102] Substitute the positive sentiment value PM and negative sentiment value NM of the original market sentiment value into the market sentiment factor MF calculation formula to obtain the market sentiment factor MF; when the market sentiment factor MF is greater than 0, the overall stock market is in an upward trend, which will enhance investor confidence; on the contrary, when the market sentiment factor MF is less than 0, it will cause investors to worry and panic;

[0103] For a specific industry, the impact of the stock market will cause the industry sentiment to move up or down, that is, the market sentiment factor MF will cause "positive deviation" and "negative deviation" to the industry sentiment. The preset sentiment deviation formula is set as:

[0104] ; Among them, r is MF, the observed value is the original industry sentiment value, and the true value is the pure industry sentiment value after eliminating the influence of market sentiment; the pure industry sentiment value includes the positive sentiment value , neutral sentiment value and negative sentiment ;

[0105] Since neutral emotions are considered a state of balance between positive and negative emotions, and the degree of hesitation between positive and negative emotions, when conducting sentiment deviation analysis, it is necessary to give restrictions on the positive and negative sentiment values ​​of the adjusted industries: if the positive sentiment value of the day increases after adjustment, it should be assumed that it all comes from the neutral sentiment value; similarly, if the negative sentiment value increases, it is also assumed that it all comes from the neutral sentiment value; in addition, assuming that the three sentiment values ​​of the two industries are 0.5, 0.2, 0.3 and 0.3, 0.2, 0.5, even if the neutral sentiment of both is 0.2, their meanings are different. If there is sentiment adjustment, the neutral sentiment of the first industry is more likely to be converted into positive sentiment. Therefore, restrictions also need to be added to the degree of conversion;

[0106] Set the preset emotion conversion rule to only convert neutral emotion to negative emotion or neutral emotion to positive emotion. and The emotional constraint formula is:

[0107] ;

[0108] ;

[0109] Determine whether MF>0 or MF<0;

[0110] If MF>0, the stock market is on an upward trend, the positive sentiment of the industry is affected, and it is falsely high, determining PI k Greater than , and the industry's negative sentiment will be lower than the real sentiment, NI k Less than ; ;

[0111] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0112] ;

[0113] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0114] ;

[0115] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0116] ;

[0117] If MF is less than 0, the stock market is in a downward trend, and the negative sentiment of the industry is inflated, which determines the PI k Less than , and the industry's positive sentiment will be lower than the real sentiment, NI k Greater than ; ;

[0118] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0119] ;

[0120] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0121] ;

[0122] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0123] ;

[0124] Combining the two pure industry sentiment values ​​under the conditions of MF>0 and MF<0, the expression of the pure industry sentiment value is obtained as follows:

[0125] .

[0126] 103, performing sentiment deviation analysis on the original individual stock sentiment value according to the original market sentiment value, and obtaining the first stock sentiment value after eliminating the influence of market sentiment;

[0127] In this embodiment, the influence of market sentiment on the sentiment value of individual stocks is eliminated, specifically:

[0128] When the market sentiment factor MF is greater than 0, , ;

[0129] When the market sentiment factor MF is less than 0, , ;

[0130] The adjusted sentiment values ​​for individual stocks are subject to the following restrictions:

[0131] , ;

[0132] The first stock sentiment value expression after removing the influence of market sentiment is:

[0133] .

[0134] 104. Perform sentiment deviation analysis on the first stock sentiment value according to the pure industry sentiment value to obtain the pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

[0135] In this embodiment, the specific process of obtaining the pure stock sentiment value after removing the market sentiment and industry sentiment is as follows:

[0136] The calculation formula for obtaining the industry sentiment factor IF is:

[0137] ;

[0138] The pure industry sentiment value and Substitute into the market sentiment factor IF calculation formula to obtain the industry sentiment factor IF;

[0139] The pure stock sentiment value is , and ;

[0140] According to the market sentiment factor IF, the sentiment deviation analysis is performed on the sentiment value of the first stock. The specific process of the sentiment deviation analysis refers to the details of step 102 and step 103. The expression of the pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment is obtained as follows:

[0141] .

[0142] The beneficial effects of the embodiments of the present invention are:

[0143] Investor sentiment data of investors within a preset time period are collected, and the investor sentiment data include original individual stock sentiment values, original industry sentiment values ​​and original market sentiment values; the original market sentiment values ​​reflect investors' investment sentiment for the entire stock market; the original industry sentiment values ​​reflect investors' investment sentiment for a specific industry in the entire stock market; the original individual stock sentiment values ​​reflect investors' investment sentiment for a specific stock in a specific industry; sentiment deviation analysis is performed on the original industry sentiment values ​​based on the original market sentiment values ​​to obtain pure industry sentiment values ​​after eliminating the influence of market sentiment; sentiment deviation analysis is performed on the original individual stock sentiment values ​​based on the original market sentiment values ​​to obtain the first stock sentiment value after eliminating the influence of market sentiment; sentiment deviation analysis is performed on the first stock sentiment value based on the pure industry sentiment value to obtain the pure individual stock sentiment value after eliminating the influence of market sentiment and industry sentiment. The influence of market sentiment on industry sentiment, as well as the influence of market sentiment and industry sentiment on individual stock sentiment are eliminated through sentiment deviation analysis technology, thereby improving the data accuracy of investor sentiment.

[0144] In combination with the investor sentiment decoupling method based on deviation analysis described in the above embodiments, the investor sentiment decoupling system based on deviation analysis is described below through an embodiment.

[0145] like Figure 2 As shown, an embodiment of the present invention provides an investor sentiment decoupling system based on deviation analysis, including:

[0146] The investor sentiment collection module 201 is used to collect the investor sentiment data of investors within a preset time period, and the investor sentiment data includes the original individual stock sentiment value, the original industry sentiment value and the original market sentiment value; the original market sentiment value reflects the investor's investment sentiment towards the entire stock market; the original industry sentiment value reflects the investor's investment sentiment towards a specific industry in the entire stock market; the original individual stock sentiment value reflects the investor's investment sentiment towards a specific stock in a specific industry; the original market sentiment value includes a positive sentiment value PM, a neutral sentiment value MM and a negative sentiment value NM; and PM+MM+NM=1; the original industry sentiment value includes a positive sentiment value PI k , neutral sentiment value MI k and negative sentiment value NI k , k represents a specific industry; and PI k +MI k +NI k =1; the original stock sentiment value includes the positive sentiment value PS m , neutral sentiment value MS m and negative sentiment value NS m , m represents a specific stock in a specific industry k; and PS m +MS m +NS m =1;

[0147] The industry sentiment decoupling module 202 is used to perform sentiment deviation analysis on the original industry sentiment value according to the original market sentiment value to obtain a pure industry sentiment value after eliminating the influence of market sentiment;

[0148] The individual stock sentiment decoupling module 203 is used to perform sentiment deviation analysis on the original individual stock sentiment value according to the original market sentiment value to obtain the first stock sentiment value after eliminating the influence of market sentiment; perform sentiment deviation analysis on the first stock sentiment value according to the pure industry sentiment value to obtain the pure individual stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

[0149] Combination Figure 2 The embodiment shown is preferably as Figure 3 As shown, in some embodiments of the present invention,

[0150] The industry sentiment decoupling module 202 includes:

[0151] Market sentiment factor calculation unit 301, configuration unit 302, judgment unit 303 and industry sentiment decoupling processing unit 304;

[0152] The market sentiment factor calculation unit 301 is used to obtain the market sentiment factor MF calculation formula:

[0153] ;

[0154] Substitute the positive sentiment value PM and the negative sentiment value NM of the original market sentiment value into the market sentiment factor MF calculation formula to obtain the market sentiment factor MF;

[0155] Configuration unit 302 is used to set the preset emotion deviation formula as:

[0156] ; Among them, r is MF, the observed value is the original industry sentiment value, and the true value is the pure industry sentiment value after eliminating the influence of market sentiment; the pure industry sentiment value includes the positive sentiment value , neutral sentiment value and negative sentiment ;

[0157] The configuration unit 302 is further configured to set a preset emotion transition rule that only a neutral emotion can be transformed into a negative emotion or a neutral emotion can be transformed into a positive emotion. and The emotional constraint formula is:

[0158] ;

[0159] ;

[0160] A judging unit 303, configured to judge whether MF>0 or MF<0;

[0161] Industry sentiment decoupling processing unit 304, used to determine PI if MF>0 k Greater than , and NI k Less than ; ;

[0162] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0163] ;

[0164] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0165] ;

[0166] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0167] ;

[0168] The industry sentiment decoupling processing unit 304 is also used to determine PI if MF < 0 k Less than , and NI k Greater than ; ;

[0169] PI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0170] ;

[0171] NI k and Substituting into the preset emotion deviation formula, we get , and is calculated by converting the formula The expression is:

[0172] ;

[0173] According to the emotional constraint formula, Expressions and The expression for the pure industry sentiment value is:

[0174] .

[0175] In summary, the beneficial effects of the investor sentiment decoupling system based on deviation analysis in the embodiment of the present invention are:

[0176] The investor sentiment collection module 201 collects the investor sentiment data of investors within a preset time period, and the investor sentiment data includes the original individual stock sentiment value, the original industry sentiment value and the original market sentiment value; the original market sentiment value reflects the investor's investment sentiment for the entire stock market; the original industry sentiment value reflects the investor's investment sentiment for a specific industry in the entire stock market; the original individual stock sentiment value reflects the investor's investment sentiment for a specific stock in a specific industry; the industry sentiment decoupling module 202 performs sentiment deviation analysis on the original industry sentiment value according to the original market sentiment value to obtain a pure industry sentiment value after eliminating the influence of market sentiment; the individual stock sentiment decoupling module 203 performs sentiment deviation analysis on the original individual stock sentiment value according to the original market sentiment value to obtain the first stock sentiment value after eliminating the influence of market sentiment; the first stock sentiment value is subjected to sentiment deviation analysis according to the pure industry sentiment value to obtain a pure individual stock sentiment value after eliminating the influence of market sentiment and industry sentiment, and the influence of market sentiment on industry sentiment and the influence of market sentiment and industry sentiment on individual stock sentiment are eliminated through the sentiment deviation analysis technology, thereby improving the data accuracy of investor sentiment.

[0177] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0178] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0179] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0180] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0181] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are included in the scope of the claims of the present invention to be approved.

Claims

1. A method for decoupling investor sentiment based on deviation analysis, characterized in that: include: Collect investor sentiment data of investors within a preset time period, wherein the investor sentiment data includes original individual stock sentiment values, original industry sentiment values, and original market sentiment values; The original market sentiment value reflects the investor's investment sentiment towards the entire stock market; the original industry sentiment value reflects the investor's investment sentiment towards a specific industry in the entire stock market; the original individual stock sentiment value reflects the investor's investment sentiment towards a specific stock in the specific industry; Performing sentiment deviation analysis on the original industry sentiment value according to the original market sentiment value to obtain a pure industry sentiment value after eliminating the influence of market sentiment; Performing sentiment deviation analysis on the original individual stock sentiment value according to the original market sentiment value to obtain the first stock sentiment value after eliminating the influence of market sentiment; The sentiment deviation analysis is performed on the first stock sentiment value according to the pure industry sentiment value to obtain a pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

2. The investor sentiment decoupling method based on deviation analysis according to claim 1 is characterized in that: The original market sentiment value includes a positive sentiment value PM, a neutral sentiment value MM and a negative sentiment value NM; and PM+MM+NM=1; The original industry sentiment value includes a positive sentiment value PI k , neutral sentiment value MI k and negative sentiment value NI k , where k represents a specific industry; and PI k +MI k +NI k =1; The original stock sentiment value includes the positive sentiment value PS m , neutral sentiment value MS m and negative sentiment value NS m , the m represents a specific stock in the specific industry k; and PS m +MS m +NS m =1.

3. The investor sentiment decoupling method based on deviation analysis according to claim 2 is characterized in that: The performing of sentiment deviation analysis on the original industry sentiment value according to the original market sentiment value to obtain a pure industry sentiment value after eliminating the influence of market sentiment includes: Calculate the market sentiment factor according to the original market sentiment value; The original industry sentiment value is subjected to sentiment deviation analysis according to the market sentiment factor to obtain a pure industry sentiment value after eliminating the influence of market sentiment.

4. The investor sentiment decoupling method based on deviation analysis according to claim 3 is characterized in that: The market sentiment factor is calculated according to the original market sentiment value, including: The calculation formula for obtaining the market sentiment factor MF is: ； Substitute the positive sentiment value PM and the negative sentiment value NM of the original market sentiment value into the market sentiment factor MF calculation formula to obtain the market sentiment factor MF.

5. The investor sentiment decoupling method based on deviation analysis according to claim 4 is characterized in that: The sentiment deviation analysis is performed on the original industry sentiment value according to the market sentiment factor to obtain a pure industry sentiment value after eliminating the influence of market sentiment, including: Set the preset sentiment bias formula as: ; Where r is the MF, the observed value is the original industry sentiment value, and the true value is the pure industry sentiment value after eliminating the influence of market sentiment; the pure industry sentiment value includes the positive sentiment value , neutral sentiment value and negative sentiment ; Set the preset emotion conversion rule to only convert neutral emotion into negative emotion or neutral emotion into positive emotion, and obtain the and stated The emotional constraint formula is: ； ； Determine whether the MF is greater than 0 or the MF is less than 0; If the MF>0, then determine the PI k Greater than , and the NI k Less than ; ; The PI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； The NI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； According to the emotion restriction formula, Expression and description The expression for the pure industry sentiment value is: ； If the MF is less than 0, the PI is determined k Less than , and the NI k Greater than ; ; The PI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； The NI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； According to the emotion restriction formula, Expression and description The expression for the pure industry sentiment value is: 。 6. The investor sentiment decoupling method based on deviation analysis according to claim 5 is characterized in that: The performing of sentiment deviation analysis on the first stock sentiment value according to the pure industry sentiment value to obtain a pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment includes: The calculation formula for obtaining the industry sentiment factor IF is: ； The pure industry sentiment value and stated Substitute into the market sentiment factor IF calculation formula to obtain the industry sentiment factor IF; The sentiment deviation analysis is performed on the first stock sentiment value according to the market sentiment factor IF to obtain a pure stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

7. An investor sentiment decoupling system based on deviation analysis, characterized in that: include: An investor sentiment collection module is used to collect investor sentiment data of investors within a preset time period, wherein the investor sentiment data includes original individual stock sentiment values, original industry sentiment values ​​and original market sentiment values; the original market sentiment values ​​reflect the investor's investment sentiment towards the entire stock market; the original industry sentiment values ​​reflect the investor's investment sentiment towards a specific industry in the entire stock market; the original individual stock sentiment values ​​reflect the investor's investment sentiment towards a specific stock in the specific industry; An industry sentiment decoupling module is used to perform sentiment deviation analysis on the original industry sentiment value according to the original market sentiment value to obtain a pure industry sentiment value after eliminating the influence of market sentiment; The individual stock sentiment decoupling module is used to perform sentiment deviation analysis on the original individual stock sentiment value according to the original market sentiment value to obtain the first stock sentiment value after eliminating the influence of market sentiment; perform sentiment deviation analysis on the first stock sentiment value according to the pure industry sentiment value to obtain the pure individual stock sentiment value after eliminating the influence of market sentiment and industry sentiment.

8. The investor sentiment decoupling system based on deviation analysis according to claim 7 is characterized in that: The original market sentiment value includes a positive sentiment value PM, a neutral sentiment value MM and a negative sentiment value NM; and PM+MM+NM=1; The original industry sentiment value includes a positive sentiment value PI k , neutral sentiment value MI k and negative sentiment value NI k , where k represents a specific industry; and PI k +MI k +NI k =1; The original stock sentiment value includes the positive sentiment value PS m , neutral sentiment value MS m and negative sentiment value NS m , the m represents a specific stock in the specific industry k; and PS m +MS m +NS m =1.

9. The investor sentiment decoupling system based on deviation analysis according to claim 8 is characterized in that: The industry sentiment decoupling module includes: Market sentiment factor calculation unit, configuration unit, judgment unit and industry sentiment decoupling processing unit; The market sentiment factor calculation unit is used to obtain the market sentiment factor MF calculation formula: ； Substituting the positive emotion value PM and the negative emotion value NM of the original market emotion value into the market emotion factor MF calculation formula to obtain the market emotion factor MF; The configuration unit is used to set the preset emotion deviation formula as: ; Where r is the MF, the observed value is the original industry sentiment value, and the true value is the pure industry sentiment value after eliminating the influence of market sentiment; the pure industry sentiment value includes the positive sentiment value , neutral sentiment value and negative sentiment ; The configuration unit is further used to set the preset emotion conversion rule to only convert the neutral emotion into the negative emotion or the neutral emotion into the positive emotion, and obtain the preset emotion conversion rule according to the preset emotion conversion rule. and stated The emotional constraint formula is: ； ； The judging unit is used to judge whether the MF is greater than 0 or the MF is less than 0; The industry sentiment decoupling processing unit is used to determine the PI if the MF>0 k Greater than , and the NI k Less than ; ; The PI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； The NI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； According to the emotion restriction formula, Expression and description The expression for the pure industry sentiment value is: ； The industry sentiment decoupling processing unit is further used to determine the PI if the MF is less than 0. k Less than , and the NI k Greater than ; ; The PI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； The NI k and stated Substituting into the preset emotion deviation formula, we get , and the above formula is converted and calculated The expression is: ； According to the emotion restriction formula, Expression and description The expression for the pure industry sentiment value is: 。

Images