Novel stock pricing algorithm based on natural language real-time processing

Abstract

The invention discloses a novel stock pricing algorithm based on natural language real-time processing, and the algorithm comprises four calculation parts: one part is risk-free return rate dynamic calculation, the risk-free return rate is the sum of the fund time value, the inflation compensation rate and the interest rate fluctuation correction term, the calculation formula is shown in the specification, the risk-free return rate represents the risk-free return rate, and the risk-free return rate is the sum of the fund time value, the inflation compensation rate and the interest rate fluctuation correction term; and representing the fund time value, the inflation compensation rate, the interest rate fluctuation correction coefficient and the interest rate fluctuation amplitude.

Description

Technical Field

[0001] This invention relates to the field of stock pricing algorithms, and in particular to a novel stock pricing algorithm based on real-time natural language processing. Background Technology

[0002] With the integration of financial market data analysis and natural language processing technologies, stock pricing, as a core application in the financial field, directly impacts the scientific nature of investment decisions in terms of the accuracy and timeliness of its results. Currently, stock pricing technology systems often use the Beta coefficient to measure the correlation between stock and market returns, and combine it with indicators such as risk-free rate of return and expected market return to calculate the expected return of stocks. Related technologies have gradually formed a technical framework based on financial data statistical analysis while also considering information processing needs.

[0003] In existing technical solutions, the calculation of the traditional Beta coefficient relies solely on regression analysis of historical stock return data and overall market return data, failing to consider the sentiment characteristics of natural language and their corresponding weights. This results in a single reflection of the correlation between stock and market returns, failing to capture the potential impact of natural language information on stock fluctuations. Consequently, the Beta coefficient's description of stock market sensitivity is limited, and it cannot comprehensively cover the influencing factors brought about by external information. Furthermore, in calculating the expected return of stocks, existing methods typically only integrate the risk-free rate of return, the traditional Beta coefficient, and the market expected return, without designing quantitative adjustment mechanisms for the potential risks that natural language information may cause. This leads to the calculation of the expected return relying solely on historical market data, failing to effectively incorporate real-time language information, and thus failing to fully consider the risk variables brought about by natural language information. This affects the rationality of stock pricing results and reduces the responsiveness of pricing results to real-time market changes. Therefore, a novel stock pricing algorithm based on real-time natural language processing is proposed. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a novel stock pricing algorithm based on real-time natural language processing, which solves the problem that it can only reflect the correlation between stock and market returns and cannot reflect the potential impact of natural language information on stock fluctuations.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a novel stock pricing algorithm based on real-time natural language processing, comprising: Dynamic calculation of risk-free rate of return: The risk-free rate of return is the sum of the time value of money, the inflation compensation rate, and the interest rate fluctuation correction term. It is the rate of return on investment excluding inflation risk and interest rate fluctuation risk, and is calculated using the following formula: in, Represents the risk-free rate of return. Represents the time value of money. Represents the inflation compensation rate. This represents the interest rate volatility correction factor. This represents the magnitude of interest rate fluctuations; Beta coefficient calculation by incorporating natural language features: The Beta coefficient, calculated using regression methods combined with real-time natural language sentiment features, reflects the correlation between stock returns and market returns, as well as the impact of natural language information on stock volatility. The calculation formula is as follows: in, The Beta coefficient represents the stock price. This represents the covariance between the return of stock A and the market return. Represents the sentiment weight of natural language. The standard deviation of market returns; and ,in, The correlation coefficient between stock A and the market. The Beta coefficient represents the standard deviation of stock a, hence it is further expressed as: Among them, the synergistic effect of each parameter jointly characterizes the correlation between stock volatility and market volatility under the influence of natural language sentiment factors; Expected rate of return calculation for different market scenarios: The expected market return is calculated based on the probability of occurrence of bull and bear market scenarios and the corresponding market returns under those scenarios. The formula is as follows: in, Represents the market's expected rate of return. This represents the probability of a bull market occurring. This represents the market return rate under a bull market scenario. This represents the probability of a bear market occurring, and , This represents the market return rate under a bear market scenario; Comprehensive calculation of expected stock returns: The expected rate of return is calculated by combining the risk-free rate of return, the beta coefficient, the market expected rate of return, and the natural language risk adjustment term. The formula is as follows: in, Represents the expected rate of return for a single stock or portfolio of stocks. This represents the natural language risk adjustment item.

[0006] Dynamic calculation of risk-free rate of return: To obtain an investment return rate that excludes inflation risk and interest rate fluctuation risk and has no other risks, the specific operation is as follows: First, clarify the components of the risk-free rate of return, including the time value of money, the inflation compensation rate, and the interest rate fluctuation correction term; second, calculate the risk-free rate of return using the following formula: in, Represents the risk-free rate of return. Represents the time value of money. Represents the inflation compensation rate. This represents the interest rate volatility correction factor. This represents the magnitude of interest rate fluctuations; Beta coefficient calculation by incorporating natural language features: The Beta coefficient, used to calculate the correlation between stock returns and market returns, and the impact of natural language information on stock volatility, is as follows: The first step is to determine the basic parameters needed to calculate the Beta coefficient, including the covariance of stock A's return and the market return, the natural language sentiment weight, and the standard deviation of the market return; the second step is to first use the formula... Calculate the covariance between the return of stock a and the market return, where, This represents the covariance between the return of stock A and the market return. The correlation coefficient between stock A and the market. The standard deviation of stock A is represented by... The standard deviation represents market returns; the third step is to calculate the Beta coefficient, which incorporates natural language features, using the following formula: Will Substituting into the above equation, the Beta coefficient can be further expressed as: in, The Beta coefficient represents the stock price. Represents the sentiment weight in natural language processing; Expected rate of return calculation for different market scenarios: This is used to calculate the expected rate of return in the market based on different market operating scenarios. The specific operation is as follows: First, identify two market scenarios: a bull market and a bear market, and define the corresponding parameters for each scenario, including the probability of a bull market occurring, the market return under a bull market, the probability of a bear market occurring, and the market return under a bear market, while satisfying the following conditions: Secondly, the expected rate of return for each market scenario is calculated using the following formula: in, Represents the market's expected rate of return. This represents the probability of a bull market occurring. This represents the market return rate under a bull market scenario. This represents the probability of a bear market occurring, and , This represents the market return rate under a bear market scenario; Comprehensive calculation of expected stock returns: The method for calculating the expected rate of return for a single stock or portfolio of stocks by considering multiple factors is as follows: First, collect the previously calculated risk-free rate of return, Beta coefficient, and expected market return, and determine the natural language risk adjustment term; second, calculate the overall expected stock return using the following formula: in, Represents the expected rate of return for a single stock or portfolio of stocks. Represents natural language risk adjustment items. Represents the risk-free rate of return. The Beta coefficient represents the stock price. This represents the market's expected rate of return.

[0007] Preferably, in the dynamic calculation of the risk-free rate of return, the interest rate fluctuation correction coefficient is used. The calculation formula is: in Represents the current market benchmark interest rate. This represents the market benchmark interest rate for the previous period.

[0008] Obtain through clear steps and And substitute into a fixed formula to calculate This ensures that the calculation process for the interest rate fluctuation correction factor is standardized and repeatable, avoiding the arbitrariness of manual calculation; it is calculated in conjunction with changes in the market benchmark interest rate. It can make It accurately reflects the fluctuations of market interest rates in different periods, provides a precise basis for the dynamic calculation of the risk-free rate of return, and makes the calculation results of the risk-free rate of return more consistent with the actual fluctuations of market interest rates.

[0009] Preferably, in the dynamic calculation of the risk-free rate of return, the interest rate fluctuation range The value is determined by obtaining real-time interest rate fluctuation monitoring data released by the central bank, and only interest rate fluctuation data released by the central bank's official channels for the same period are selected.

[0010] Limiting interest rate fluctuation range The values ​​are sourced exclusively from official central bank channels. Leveraging the central bank's authority as the core financial regulatory institution, this ensures the authenticity and reliability of the obtained interest rate fluctuation data, avoiding potential errors or false information arising from the use of unofficial data. The impact of the selected values; at the same time, data from the same period should be selected to ensure... Matching the risk-free rate of return calculation period, the dynamic calculation of the risk-free rate of return can be based on accurate interest rate fluctuation information during the same period, thereby improving the accuracy of the risk-free rate of return calculation.

[0011] Preferably, in the calculation of the Beta coefficient that integrates natural language features, the natural language sentiment weight... The calculation formula is: in Represents the natural language sentiment score. This was determined by semantic analysis of real-time stock-related news and industry research reports. The semantic analysis process involves extracting positive and negative words related to stock returns from the text and counting their frequencies.

[0012] By collecting real-time stock-related news and industry research reports, and focusing on semantic analysis of words related to stock returns, we can obtain... ,make It can accurately reflect the market's sentiment towards stock returns; based on Calculating Natural Language Sentiment Weights By incorporating natural language features into the calculation of the Beta coefficient, the traditional Beta coefficient, which relies solely on financial data, is broken. This allows the Beta coefficient to more comprehensively reflect the impact of market sentiment on the risk-return characteristics of stocks, thereby enhancing the comprehensiveness and practicality of the Beta coefficient.

[0013] Preferably, in the calculation of the Beta coefficient that integrates natural language features, it is used to calculate the natural language sentiment score. The text must meet the following conditions: the time of publication of the text is no more than 72 hours from the current calculation time; the text is published by a licensed financial institution or authoritative financial media; and the text content contains specific financial data of stocks or statements of industry policies.

[0014] The calculation was analyzed from three dimensions: time, publisher, and content. The text is filtered, and a 72-hour time limit ensures the timeliness of the text information, avoiding the use of outdated information. The restriction on publication by licensed financial institutions and authoritative financial media ensures the professionalism and credibility of the text information, reducing the influence of non-professional information. Interference with accuracy; content restrictions including specific financial data on stocks or statements of industry policies ensure a high correlation between the text and stock returns, making... It can more accurately reflect the sentiment tendency related to stock returns, thus providing high-quality sentiment data support for the calculation of the Beta coefficient.

[0015] Preferably, in the calculation of the expected rate of return for each market scenario, the probability of a bull market scenario occurring is considered. The calculation formula is: in This represents the percentage of days in which the market index rose over the past 30 days. This represents the percentage of days in which the market trading volume over the past 30 days is higher than the average trading volume over the past 180 days.

[0016] Combining the percentage of days with market index increases over the past 30 days The percentage of days with market trading volume exceeding the average trading volume of the past 180 days in the past 30 days. To calculate It takes into account both the rise in market prices and the performance of market trading activity, making... The computational dimensions are more comprehensive; through fixed weight allocation, 0.6 It accounts for 0.4, and is used for calculation to ensure that The consistency and objectivity of the calculation process avoids calculation deviations caused by human adjustment of weights, allowing... It can more reasonably reflect the probability of a bull market scenario and provide an accurate basis for calculating the expected rate of return in different market scenarios.

[0017] Preferably, in the calculation of the expected rate of return for each market scenario, the market return rate under a bull market scenario is determined. The basis for this includes the cumulative increase in the market index over the past three months, the average increase in industry sectors, and the macroeconomic prosperity index. The value of should be determined by combining the arithmetic mean of the above three indicators.

[0018] The three indicators selected are the cumulative increase of the market index over the past three months, the average increase of industry sectors, and the macroeconomic prosperity index. The cumulative increase in the market index over the past three months reflects the overall price return of the market, the average increase of industry sectors reflects the return performance at the industry level, and the macroeconomic prosperity index reflects the impact of the macroeconomic environment on market returns. The combination of these three factors... The determination of the value can comprehensively consider factors at three levels: market, industry, and macroeconomics; it is determined by calculating the arithmetic mean of the three indicators. This ensures that all indicators are accurate and reliable. The balanced impact avoids the dominance of a single indicator. The bias caused by the choice of values ​​affects market returns in a bull market. It is more in line with the actual market returns.

[0019] Preferably, in the comprehensive calculation of the expected stock return, the natural language risk adjustment item... The calculation formula is: in Represents the natural language sentiment score. The Beta coefficient represents the stock's value, and when... When it is a positive value When it is a positive value, When it is negative It is a negative value.

[0020] Natural Language Sentiment Score and the Beta coefficient of stocks Integrating natural language risk adjustment items In the calculation, This reflects the impact of natural market sentiment on stocks. This reflects the systematic risk level of a stock; the combination of the two makes... It can simultaneously consider the adjusting effects of emotional factors and systematic risk factors on expected stock returns; according to Determining the positive and negative aspects The positive and negative properties make The direction aligns with market sentiment; when market sentiment is positive... A positive value has a positive adjusting effect on the expected return of stocks; when market sentiment is negative, A negative value serves as a negative adjustment, allowing the comprehensive calculation of expected stock returns to more accurately reflect the impact of natural language risk factors.

[0021] Preferably, in the comprehensive calculation of the expected stock return, the natural language risk adjustment item... Update cycle and natural language sentiment score The update cycle is consistent, and each update... The stock's Beta coefficient needs to be recalculated simultaneously. and market expected rate of return .

[0022] make Update cycle and The update cycle is consistent to ensure It can be updated in a timely manner according to changes in the natural language sentiment of the market, avoiding Out of touch with current market sentiment; in updating Time synchronization recalculation and ,because It will change with changes in stock returns and market returns. It will be adjusted according to changes in the overall market situation; synchronized updates ensure... The calculation is based on the latest and Data, to avoid due to and Failure to update in time The calculation basis is outdated, thus ensuring that all parameters relied upon for the comprehensive calculation of expected stock returns are up-to-date, thereby improving the timeliness and accuracy of the expected stock returns calculation results.

[0023] Preferably, in the calculation of the Beta coefficient incorporating natural language features, the covariance between the return of stock a and the market return is calculated. When doing so, a weighted allocation between historical and real-time data needs to be introduced, and the calculation formula is as follows: in Represents the covariance calculated based on historical data. Represents the weight of historical data. Represents the covariance calculated based on real-time data. Represents real-time data weights, and .

[0024] In calculation The introduction of weighted historical and real-time data allows for the assessment of the long-term stability of the relationship between stock returns and market returns, while real-time data reflects recent trends in this relationship. Combining the two data points enables… The calculation takes into account both long-term stability and timely reflection of recent dynamic changes, avoiding the problems of failing to capture recent relationship changes due to relying solely on historical data, or being overly affected by short-term fluctuations due to relying solely on real-time data; by setting... The weighting rules ensure that The rationality and logic of the calculation enable the covariance results to more accurately reflect the correlation between stock A's returns and market returns, providing precise covariance data support for the calculation of Beta coefficients that incorporate natural language features.

[0025] In summary, compared with existing technologies, this invention provides a novel stock pricing algorithm based on real-time natural language processing, which has the following beneficial effects: The Beta coefficient calculation in this invention integrates natural language features. By combining regression methods with real-time natural language sentiment features and introducing natural language sentiment weights, it breaks through the limitation of the traditional Beta coefficient, which only reflects the correlation between stock and market returns. It can simultaneously reflect the impact of natural language information on stock fluctuations, making the Beta coefficient more comprehensively reflect the market sensitivity of stocks and the impact of external information. In the comprehensive calculation of expected stock returns, a natural language risk adjustment term is added on the basis of combining the risk-free rate of return, the improved Beta coefficient, and the expected rate of return in different market scenarios. This achieves a quantitative adjustment of the potential risks brought by natural language information, making the calculation of expected returns more comprehensively integrate market data and real-time language information, and improving the rationality and timeliness of stock pricing. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the algorithm of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Please see Figure 1 This invention provides a technical solution: a novel stock pricing algorithm based on real-time natural language processing, comprising: Dynamic calculation of risk-free rate of return: The risk-free rate of return is the sum of the time value of money, the inflation compensation rate, and the interest rate fluctuation correction term. It is the rate of return on investment excluding inflation risk and interest rate fluctuation risk, and is calculated using the following formula: in, Represents the risk-free rate of return. Represents the time value of money. Represents the inflation compensation rate. This represents the interest rate volatility correction factor. This represents the magnitude of interest rate fluctuations; Beta coefficient calculation by incorporating natural language features: The Beta coefficient, calculated using regression methods combined with real-time natural language sentiment features, reflects the correlation between stock returns and market returns, as well as the impact of natural language information on stock volatility. The calculation formula is as follows: in, The Beta coefficient represents the stock price. This represents the covariance between the return of stock A and the market return. Represents the sentiment weight of natural language. The standard deviation of market returns; and ,in, The correlation coefficient between stock A and the market. The Beta coefficient represents the standard deviation of stock a, hence it is further expressed as: Among them, the synergistic effect of each parameter jointly characterizes the correlation between stock volatility and market volatility under the influence of natural language sentiment factors; Expected rate of return calculation for different market scenarios: The expected market return is calculated based on the probability of occurrence of bull and bear market scenarios and the corresponding market returns under those scenarios. The formula is as follows: in, Represents the market's expected rate of return. This represents the probability of a bull market occurring. This represents the market return rate under a bull market scenario. This represents the probability of a bear market occurring, and , This represents the market return rate under a bear market scenario; Comprehensive calculation of expected stock returns: The expected rate of return is calculated by combining the risk-free rate of return, the beta coefficient, the market expected rate of return, and the natural language risk adjustment term. The formula is as follows: in, Represents the expected rate of return for a single stock or portfolio of stocks. This represents the natural language risk adjustment item.

[0029] Dynamic calculation of risk-free rate of return: To obtain an investment return rate that excludes inflation risk and interest rate fluctuation risk and has no other risks, the specific operation is as follows: First, clarify the components of the risk-free rate of return, including the time value of money, the inflation compensation rate, and the interest rate fluctuation correction term; second, calculate the risk-free rate of return using the following formula: in, Represents the risk-free rate of return. Represents the time value of money. Represents the inflation compensation rate. This represents the interest rate volatility correction factor. This represents the magnitude of interest rate fluctuations; Beta coefficient calculation by incorporating natural language features: The Beta coefficient, used to calculate the correlation between stock returns and market returns, and the impact of natural language information on stock volatility, is as follows: The first step is to determine the basic parameters needed to calculate the Beta coefficient, including the covariance of stock A's return and the market return, the natural language sentiment weight, and the standard deviation of the market return; the second step is to first use the formula... Calculate the covariance between the return of stock a and the market return, where, This represents the covariance between the return of stock A and the market return. The correlation coefficient between stock A and the market. The standard deviation of stock A is represented by... The standard deviation represents market returns; the third step is to calculate the Beta coefficient, which incorporates natural language features, using the following formula: Will Substituting into the above equation, the Beta coefficient can be further expressed as: in, The Beta coefficient represents the stock price. Represents the sentiment weight in natural language processing; Expected rate of return calculation for different market scenarios: This is used to calculate the expected rate of return in the market based on different market operating scenarios. The specific operation is as follows: First, identify two market scenarios: a bull market and a bear market, and define the corresponding parameters for each scenario, including the probability of a bull market occurring, the market return under a bull market, the probability of a bear market occurring, and the market return under a bear market, while satisfying the following conditions: Secondly, the expected rate of return for each market scenario is calculated using the following formula: in, Represents the market's expected rate of return. This represents the probability of a bull market occurring. This represents the market return rate under a bull market scenario. This represents the probability of a bear market occurring, and , This represents the market return rate under a bear market scenario; Comprehensive calculation of expected stock returns: The method for calculating the expected rate of return for a single stock or portfolio of stocks by considering multiple factors is as follows: First, collect the previously calculated risk-free rate of return, Beta coefficient, and expected market return, and determine the natural language risk adjustment term; second, calculate the overall expected stock return using the following formula: in, Represents the expected rate of return for a single stock or portfolio of stocks. Represents natural language risk adjustment items. Represents the risk-free rate of return. The Beta coefficient represents the stock price. Represents the market's expected rate of return; The calculation of the risk-free rate of return incorporates an interest rate volatility correction item, which is reflected by the product of the interest rate volatility correction coefficient and the interest rate volatility amplitude. Compared with the calculation method that only considers the time value of money and the inflation compensation rate, it can more comprehensively eliminate the impact of interest rate volatility risk on investment returns, making the calculated risk-free rate of return more closely reflect the returns of actual risk-free investment scenarios. The Beta coefficient is calculated by introducing natural language sentiment weights and combining parameters such as the covariance of stock returns and market returns and the standard deviation of market returns. It can reflect the degree of correlation between stock returns and market returns, while also reflecting the impact of natural language information on stock volatility, making the stock volatility-related characteristics reflected by the Beta coefficient more comprehensive. The calculation of the expected market return rate is based on two common market scenarios: bull market and bear market. It is calculated by combining the probability of occurrence of the two scenarios and the market return rate under the corresponding scenarios. This is consistent with the characteristics of different scenarios in actual market operation, so that the calculated expected market return rate is more in line with the actual market situation. The calculation of expected stock returns integrates risk-free rate of return, Beta coefficient incorporating natural language features, market expected rate of return calculated in different scenarios, and introduces a natural language risk adjustment term. This allows for the consideration of multiple factors such as risk-free return, correlation with market returns, differences in market scenarios, and natural language risk, making the calculated expected stock returns more comprehensive and reasonable.

[0030] In the dynamic calculation of the risk-free rate of return, the interest rate volatility correction factor is used. The calculation formula is: in Represents the current market benchmark interest rate. This represents the market benchmark interest rate for the previous period.

[0031] The first step is to clarify the need for dynamic calculation of the risk-free rate of return and determine the interest rate volatility correction factor that needs to be calculated. The second step is to collect the current market benchmark interest rate. The first step is to obtain this data through authoritative channels such as financial data platforms and central bank release channels; the second step is to collect the market benchmark interest rate for the previous period. To ensure that this data is consistent with the current market benchmark interest rate. The statistical periods are consistent to ensure data comparability; the fourth step is to obtain the data... and Substitute into the formula First calculate and The absolute value of the difference, then multiply that absolute value by... Finally, use Subtracting the result yields the interest rate volatility correction factor. ; Obtain through clear steps and And substitute into a fixed formula to calculate This ensures that the calculation process for the interest rate fluctuation correction factor is standardized and repeatable, avoiding the arbitrariness of manual calculation; it is calculated in conjunction with changes in the market benchmark interest rate. It can make It accurately reflects the fluctuations of market interest rates in different periods, provides a precise basis for the dynamic calculation of the risk-free rate of return, and makes the calculation results of the risk-free rate of return more consistent with the actual fluctuations of market interest rates.

[0032] In the dynamic calculation of the risk-free rate of return, the magnitude of interest rate fluctuations The value is determined by obtaining real-time interest rate fluctuation monitoring data released by the central bank, and only interest rate fluctuation data released by the central bank's official channels for the same period are selected.

[0033] The first step, in the dynamic calculation of the risk-free rate of return, is to determine the required interest rate fluctuation range. The second step is to determine the value of the interest rate fluctuation range. The first step involves using real-time interest rate fluctuation monitoring data released by the central bank, excluding interest rate fluctuation data from other non-central bank channels. The second step involves obtaining interest rate fluctuation data for the same period as the current risk-free rate of return calculation cycle from official channels such as the central bank's official website and official statistical reports. The third step involves determining the obtained interest rate fluctuation data from official central bank channels as the interest rate fluctuation range. The possible values ​​of ; Limiting interest rate fluctuation range The values ​​are sourced exclusively from official central bank channels. Leveraging the central bank's authority as the core financial regulatory institution, this ensures the authenticity and reliability of the obtained interest rate fluctuation data, avoiding potential errors or false information arising from the use of unofficial data. The impact of the selected values; at the same time, data from the same period should be selected to ensure... Matching the risk-free rate of return calculation period, the dynamic calculation of the risk-free rate of return can be based on accurate interest rate fluctuation information during the same period, thereby improving the accuracy of the risk-free rate of return calculation.

[0034] In the calculation of the Beta coefficient by incorporating natural language features, the natural language sentiment weight is... The calculation formula is: in Represents the natural language sentiment score. This was determined by semantic analysis of real-time stock-related news and industry research reports. The semantic analysis process involves extracting positive and negative words related to stock returns from the text and counting their frequencies.

[0035] The first step, in the Beta coefficient calculation work that integrates natural language features, is to initiate the natural language sentiment weighting. The process involves four steps: first, the calculation of the data; second, collecting real-time stock-related news and industry research reports to ensure the text covers the latest information on the current stock market and corresponding industries; third, performing semantic analysis on the collected text, focusing on extracting positive and negative words related to stock returns and counting their frequencies; and fourth, calculating the natural language sentiment score based on the frequency of positive and negative words. Fifth step, Substitute into the formula The natural language sentiment weights were calculated. ; By collecting real-time stock-related news and industry research reports, and focusing on semantic analysis of words related to stock returns, we can obtain... ,make It can accurately reflect the market's sentiment towards stock returns; based on Calculating Natural Language Sentiment Weights By incorporating natural language features into the calculation of the Beta coefficient, the traditional Beta coefficient, which relies solely on financial data, is broken. This allows the Beta coefficient to more comprehensively reflect the impact of market sentiment on the risk-return characteristics of stocks, thereby enhancing the comprehensiveness and practicality of the Beta coefficient.

[0036] In the calculation of the Beta coefficient, which incorporates natural language features, it is used to calculate the natural language sentiment score. The text must meet the following conditions: the time of publication of the text is no more than 72 hours from the current calculation time; the text is published by a licensed financial institution or authoritative financial media; and the text content contains specific financial data of stocks or statements of industry policies.

[0037] The first step, in the calculation of the Beta coefficient by incorporating natural language features, is to address the issue of calculating the natural language sentiment score. The text screening process is as follows: First, the text is filtered. Second, the publication time of the text is checked, and texts published within 72 hours of the current calculation time are selected, while texts published too long ago are excluded. Third, the publishing entity of the text is verified, and texts published by licensed financial institutions or authoritative financial media are selected, while texts published by irregular entities are excluded. Fourth, the content of the text is reviewed, and texts containing specific financial data on stocks or statements of industry policies are selected, while texts with empty content or low correlation to stock returns are excluded. Fifth, texts that meet all three conditions above are selected for calculation. Valid text; The calculation was analyzed from three dimensions: time, publisher, and content. The text is filtered, and a 72-hour time limit ensures the timeliness of the text information, avoiding the use of outdated information. The restriction on publication by licensed financial institutions and authoritative financial media ensures the professionalism and credibility of the text information, reducing the influence of non-professional information. Interference with accuracy; content restrictions including specific financial data on stocks or statements of industry policies ensure a high correlation between the text and stock returns, making... It can more accurately reflect the sentiment tendency related to stock returns, thus providing high-quality sentiment data support for the calculation of the Beta coefficient.

[0038] In calculating the expected rate of return for different market scenarios, the probability of a bull market scenario occurring... The calculation formula is: in This represents the percentage of days in which the market index rose over the past 30 days. This represents the percentage of days in which the market trading volume over the past 30 days is higher than the average trading volume over the past 180 days.

[0039] The first step is to calculate the probability of a bull market scenario in the process of calculating the expected rate of return for different market scenarios. The calculation process is as follows: The second step is to count the number of days the market index rose in the past 30 days, and divide the number of days of increase by 30 to obtain the percentage of days the market index rose in the past 30 days. The third step is to calculate the average market trading volume over the past 180 days, then count the number of days in the past 30 days where the market trading volume was higher than this average. Divide the number of days higher than the average by 30 to obtain the percentage of days in the past 30 days where the market trading volume was higher than the average trading volume over the past 180 days. Fourth step, and Substitute into the formula First calculate separately and Then add the two results together to get the probability of a bull market occurring. ; Combining the percentage of days with market index increases over the past 30 days The percentage of days with market trading volume exceeding the average trading volume of the past 180 days in the past 30 days. To calculate It takes into account both the rise in market prices and the performance of market trading activity, making... The computational dimensions are more comprehensive; through fixed weight allocation, 0.6 It accounts for 0.4, and is used for calculation to ensure that The consistency and objectivity of the calculation process avoids calculation deviations caused by human adjustment of weights, allowing... It can more reasonably reflect the probability of a bull market scenario and provide an accurate basis for calculating the expected rate of return in different market scenarios.

[0040] In calculating the expected rate of return in different market scenarios, we determine the market return rate under a bull market scenario. The basis for this includes the cumulative increase in the market index over the past three months, the average increase in industry sectors, and the macroeconomic prosperity index. The value of should be determined by combining the arithmetic mean of the above three indicators.

[0041] The first step, in calculating the expected rate of return for different market scenarios, is to initiate the market yield calculation once a bull market is identified. The process involves five steps: first, calculating the cumulative increase in the market index over the past three months; second, obtaining data from financial statistics databases and information released by stock exchanges; third, collecting data on the increases of various industry sectors and calculating the average increase of each sector; fourth, obtaining the macroeconomic prosperity index from authoritative sources such as macroeconomic research institutions and government statistical departments; and fifth, calculating the arithmetic mean of the three indicators—the cumulative increase in the market index over the past three months, the average increase of each industry sector, and the macroeconomic prosperity index—and determining this average as the market return under a bull market scenario. The possible values ​​of ; The three indicators selected are the cumulative increase of the market index over the past three months, the average increase of industry sectors, and the macroeconomic prosperity index. The cumulative increase in the market index over the past three months reflects the overall price return of the market, the average increase of industry sectors reflects the return performance at the industry level, and the macroeconomic prosperity index reflects the impact of the macroeconomic environment on market returns. The combination of these three factors... The determination of the value can comprehensively consider factors at three levels: market, industry, and macroeconomics; it is determined by calculating the arithmetic mean of the three indicators. This ensures that all indicators are accurate and reliable. The balanced impact avoids the dominance of a single indicator. The bias caused by the choice of values ​​affects market returns in a bull market. It is more in line with the actual market returns.

[0042] In the comprehensive calculation of expected stock returns, the natural language risk adjustment item... The calculation formula is: in Represents the natural language sentiment score. The Beta coefficient represents the stock's value, and when... When it is a positive value When it is a positive value, When it is negative It is a negative value.

[0043] The first step is to implement a natural language risk adjustment item during the comprehensive calculation of expected stock returns. The first step is the computation; the second step is to obtain the calculated natural language sentiment score. and the Beta coefficient of stocks The third step is to... and Substitute into the formula First calculate The result, then calculate Finally, multiply the two results to obtain a preliminary result. Value; Step 4, based on Positive and negative judgments The positive and negative values, if If it is positive, then determine If it is a positive value; If it is negative, then it is determined. It is a negative value; Natural Language Sentiment Score and the Beta coefficient of stocks Integrating natural language risk adjustment items In the calculation, This reflects the impact of natural market sentiment on stocks. This reflects the systematic risk level of a stock; the combination of the two makes... It can simultaneously consider the adjusting effects of emotional factors and systematic risk factors on expected stock returns; according to Determining the positive and negative aspects The positive and negative properties make The direction aligns with market sentiment; when market sentiment is positive... A positive value has a positive adjusting effect on the expected return of stocks; when market sentiment is negative, A negative value serves as a negative adjustment, allowing the comprehensive calculation of expected stock returns to more accurately reflect the impact of natural language risk factors.

[0044] In the comprehensive calculation of expected stock returns, the natural language risk adjustment item... Update cycle and natural language sentiment score The update cycle is consistent, and each update... The stock's Beta coefficient needs to be recalculated simultaneously. and market expected rate of return .

[0045] The first step is to clarify the natural language risk adjustment item in the comprehensive calculation of expected stock returns. The update rules; the second step is to determine the natural language sentiment score. The update cycle will The update cycle is set to be consistent with The update cycle remains consistent; the third step is to ensure that the update cycle is consistent with the previous one. During the update cycle, first obtain the latest natural language sentiment score. Fourth step: Simultaneously recalculate the stock's Beta coefficient. By collecting the latest stock return data and market return data, and following the calculation logic of the Beta coefficient, the process is completed. The fifth step is to simultaneously recalculate the expected market return. Based on the latest market data and scenario-specific calculation logic, the following conclusions are drawn. Step 6, the latest Recalculated and Substitution The calculation formula is used to obtain the updated version. value; make Update cycle and The update cycle is consistent to ensure It can be updated in a timely manner according to changes in the natural language sentiment of the market, avoiding Out of touch with current market sentiment; in updating Time synchronization recalculation and ,because It will change with changes in stock returns and market returns. It will be adjusted according to changes in the overall market situation; synchronized updates ensure... The calculation is based on the latest and Data, to avoid due to and Failure to update in time The calculation basis is outdated, thus ensuring that all parameters relied upon for the comprehensive calculation of expected stock returns are up-to-date, thereby improving the timeliness and accuracy of the expected stock returns calculation results.

[0046] In the calculation of the Beta coefficient incorporating natural language features, the covariance between the return of stock a and the market return is calculated. When doing so, a weighted allocation between historical and real-time data needs to be introduced, and the calculation formula is as follows: in Represents the covariance calculated based on historical data. Represents the weight of historical data. Represents the covariance calculated based on real-time data. Represents real-time data weights, and .

[0047] The first step, in calculating the Beta coefficient by incorporating natural language features, is to start by calculating the covariance between the return of stock a and the market return. The first step is to perform calculations; the second step is to collect historical return data for stock A and the market, and calculate the covariance based on the historical data according to the covariance calculation method. The third step is to collect real-time return data for stock A and the market, and then calculate the covariance based on the real-time data using the same covariance calculation method. The fourth step is to determine the weights of historical data. and real-time data weights ,make sure The determination of weights needs to take into account factors such as the stability of the stock market and the validity of the data; the fifth step is to... , , and Substitute into the formula First calculate separately and Then, by adding the two results together, we obtain the covariance between the return of stock a and the market return. ; In calculation The introduction of weighted historical and real-time data allows for the assessment of the long-term stability of the relationship between stock returns and market returns, while real-time data reflects recent trends in this relationship. Combining the two data points enables… The calculation takes into account both long-term stability and timely reflection of recent dynamic changes, avoiding the problems of failing to capture recent relationship changes due to relying solely on historical data, or being overly affected by short-term fluctuations due to relying solely on real-time data; by setting... The weighting rules ensure that The rationality and logic of the calculation enable the covariance results to more accurately reflect the correlation between stock A's returns and market returns, providing precise covariance data support for the calculation of Beta coefficients that incorporate natural language features.

[0048] This plan: Dynamic calculation of risk-free rate of return: clearly defines the components of the risk-free rate of return, including the time value of money. Inflation compensation rate With interest rate fluctuation correction term; based on formula Calculate the risk-free rate of return ,in Represents the risk-free rate of return. This represents the interest rate volatility correction factor. Represents the magnitude of interest rate fluctuations; calculates the interest rate fluctuation correction factor. Collect current market benchmark interest rates This data is obtained through authoritative channels such as financial data platforms and central bank releases; the market benchmark interest rate for the previous period is collected. Ensure that the data is consistent with The statistical period is consistent; and Substitute into the formula First calculate and The absolute value of the difference, then multiply that absolute value by... Finally, use Subtracting the result, we get Determine the range of interest rate fluctuations. Values: Clear The data source is real-time interest rate fluctuation monitoring data released by the People's Bank of China (PBOC), excluding interest rate fluctuation data from other non-PBOC channels; interest rate fluctuation data for the same period as the current risk-free rate of return calculation period is obtained from official channels such as the PBOC's official website and official statistical reports; the obtained interest rate fluctuation data released by official PBOC channels for the same period is determined as... The possible values ​​of ; Beta coefficient calculation incorporating natural language features: Determining the basic parameters required for calculating the Beta coefficient, including the covariance between the return of stock a and the market return. Natural Language Sentiment Weight Standard deviation of market returns ;calculate Collect historical return data for stock A and the market, and derive the covariance based on the historical data using the covariance calculation method. Collect real-time return data for stock A and the market, and derive the covariance based on the real-time data using the covariance calculation method. Determine the weights of historical data. and real-time data weights ,make sure ;Will , , and Substitute into the formula First calculate separately and Then add the two results together to get ; Filtering for calculating natural language sentiment scores The text is processed as follows: The publication time of the text is checked, and texts published within 72 hours of the current calculation time are selected; the publishing entity of the text is verified, and texts published by licensed financial institutions or authoritative financial media are selected; the content of the text is reviewed, and texts containing specific financial data of stocks or statements of industry policies are selected; texts that meet all three conditions are selected for calculation. Valid text; calculation Semantic analysis is performed on the selected valid text to extract positive and negative words related to stock returns; the frequency of occurrence of positive and negative words is counted, and the following calculations are made based on the frequency: ; Calculate natural language sentiment weights :Will Substitute into the formula Calculations yielded ; Calculate the Beta coefficient :Will , and Substitute into the formula Or substitute into the formula for further expression. ,in The correlation coefficient between stock A and the market. The standard deviation of stock A is calculated as follows: ; Calculation of expected return in different market scenarios: This involves identifying two market scenarios: a bull market and a bear market, along with corresponding parameters, including the probability of a bull market occurring. Market yield under a bull market scenario The probability of a bear market. Market returns in a bear market And satisfy ;calculate Calculate the number of days the market index rose in the past 30 days, divide the number of days of increase by 30, and get the percentage of days the market index rose in the past 30 days. Calculate the average market trading volume over the past 180 days, then count the number of days in the past 30 days where the market trading volume was higher than this average. Divide the number of days higher than the average by 30 to obtain the percentage of days in the past 30 days where the market trading volume was higher than the average trading volume over the past 180 days. ;Will and Substitute into the formula First calculate separately and Then add the two results together to get ;Sure The process involves: obtaining cumulative market index gains over the past three months through financial statistics databases and information released by stock exchanges; collecting gains data from various industry sectors and calculating the average gain for each sector; obtaining macroeconomic prosperity indices from authoritative sources such as macroeconomic research institutions and government statistical departments; and calculating the arithmetic mean of these three indicators—cumulative market index gains over the past three months, average sector gains, and macroeconomic prosperity index—and determining this average as the benchmark. Calculate the market expected rate of return. :Will , , and Substitute into the formula Calculations yielded ; Comprehensive calculation of expected stock returns: This involves collecting previously calculated risk-free rates of return. Beta coefficient Compared with market expected return At the same time, natural language risk adjustment items were determined. Related Natural Language Sentiment Scores ;calculate :Will and Substitute into the formula First calculate The result, then calculate Finally, multiply the two results to obtain a preliminary result. Value; according to Positive and negative judgments The positive and negative values, if If it is positive, then determine If it is a positive value; If it is negative, then it is determined. Negative values; set Update cycle and related parameters for synchronous updates: Determine The update cycle will The update cycle is set to be consistent with The update cycle remains consistent; when it reaches During the update cycle, obtain the latest ; Synchronous recalculation By collecting the latest stock return data and market return data, and following the calculation logic of the Beta coefficient, the process is completed. Recalculation; Synchronous recalculation Based on the latest market data and scenario-specific calculation logic, the following conclusions are drawn. The latest Recalculated and Substitution The calculation formula is used to obtain the updated version. Value; Calculate the expected return on stocks :Will , , and Substitute into the formula Calculate the expected rate of return for a single stock or portfolio of stocks. .

[0049] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0050] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A new algorithm for real-time processing of stock pricing based on natural language, characterized in that, Comprising: Dynamic calculation of risk-free return rate: The risk-free return rate is the sum of the time value of money, the inflation compensation rate and the interest rate fluctuation correction term. It is the investment yield rate excluding the risks of inflation and interest rate fluctuations. The calculation formula is: wherein, represents a risk-free return rate, represents a capital time value, represents an inflation compensation rate, represents a rate fluctuation correction coefficient, represents a rate fluctuation range; Calculation of Beta coefficient combined with natural language features: The Beta coefficient is calculated by combining real-time natural language sentiment features through regression method, which is used to reflect the correlation between stock returns and market returns and the influence of natural language information on stock volatility. The calculation formula is: wherein, a Beta coefficient representing a stock, a covariance between the return of stock a and the market return, a natural language sentiment weight, a standard deviation of the market return; and wherein, a correlation coefficient between stock a and the market, a standard deviation of stock a, so that the Beta coefficient is further represented as: Wherein, the synergistic effect of each parameter jointly depicts the volatility correlation characteristics of the stock relative to the market under the influence of natural language sentiment factors; Calculation of market expected return rate in different scenarios: The market expected return rate is calculated based on the occurrence probability of bull market and bear market and the market return rate in the corresponding scenario. The calculation formula is: wherein, represents the market expected return rate, represents the probability of a bullish market scenario, represents the market return rate in a bullish market scenario, represents the probability of a bearish market scenario, and , represents the market return rate in a bearish market scenario; Comprehensive calculation of stock expected return rate: The expected return rate is calculated by combining the risk-free return rate, the Beta coefficient, the market expected return rate and the natural language risk adjustment term. The formula is: wherein, represents the expected return rate of a single stock or a portfolio of stocks, represents the natural language risk adjustment term.

2. A new stock pricing algorithm based on natural language real-time processing according to claim 1, characterized in that: In the dynamic calculation of the risk-free return rate, the interest rate fluctuation correction coefficient is calculated by the following formula: wherein represents the market benchmark interest rate for the current period, represents the market benchmark interest rate for the previous period.

3. A new stock pricing algorithm based on natural language real-time processing according to claim 2, characterized in that: In the dynamic calculation of the risk-free rate of return, the interest rate fluctuation range The value is determined by obtaining real-time interest rate fluctuation monitoring data released by the central bank, and only interest rate fluctuation data released by the central bank's official channels for the same period are selected.

4. A new stock pricing algorithm based on natural language real-time processing according to claim 1, characterized in that: In the calculation of the Beta coefficient of the fused natural language feature, the calculation formula of the natural language sentiment weight is: ​ wherein represent natural language sentiment scores, By performing semantic analysis on real-time stock-related news and industry research texts, and the semantic analysis process extracts positive and negative words related to stock returns in the text and counts the frequency.

5. A new stock pricing algorithm based on natural language real-time processing according to claim 4, characterized in that: In the Beta coefficient calculation of the fused natural language feature, the text used to calculate the natural language sentiment score must meet the following conditions: the text publishing time is not more than 72 hours from the current calculation time, the text publishing subject is a licensed financial institution or an authoritative financial media, and the text content contains specific financial data of stocks or industry policy expressions.

6. A new stock pricing algorithm based on natural language real-time processing according to claim 1, characterized in that: The probability of the occurrence of the bull market scene in the calculation of the expected return rate of the sub-scene market The calculation formula is: wherein represents the proportion of days in the last 30 days when the market index rose, represents the proportion of days in the last 30 days when the market trading volume was higher than the average trading volume in the last 180 days.

7. A new algorithm for real-time stock pricing based on natural language processing according to claim 6, characterized in that: In the calculation of expected market returns for different scenarios, the market return rate under a bull market scenario is determined. The basis for this includes the cumulative increase in the market index over the past three months, the average increase in industry sectors, and the macroeconomic prosperity index. The value of should be determined by combining the arithmetic mean of the above three indicators.

8. A new algorithm for stock pricing based on natural language real-time processing according to claim 1, characterized in that: The natural language risk adjustment item in the comprehensive calculation of the expected return rate of the stock The calculation formula is: wherein represents a natural language sentiment score, represents a Beta coefficient for a stock, and when is positive is positive, when is negative is negative.

9. A new stock pricing algorithm based on natural language real-time processing according to claim 8, characterized in that: In the comprehensive calculation of the expected stock return, the natural language risk adjustment item... Update cycle and natural language sentiment score The update cycle is consistent, and each update... The stock's Beta coefficient needs to be recalculated simultaneously. and market expected rate of return .

10. The stock pricing novel algorithm based on natural language real-time processing according to claim 1, characterized in that: In the calculation of the Beta coefficient of the fused natural language features, the covariance of the return of stock a and the market return is calculated When the historical data and the real-time data are introduced, the weight distribution is calculated, and the calculation formula is: wherein represents a covariance calculated based on historical data, represents a weight of historical data, represents a covariance calculated based on real-time data, represents a weight of real-time data, and .

Images