Stock staring and investment strategy optimization method and system in intelligent investment advisor system

Abstract

The invention relates to the technical field of financial management, and discloses a stock staring and investment strategy optimization method and system in an intelligent investment adviser system, and the method comprises the steps: analyzing a financing scene classification label corresponding to a target stock, collecting the domain volatility and liquidity index of the target stock, and generating a putting combination distribution echelon of the target stock; calculating a fluctuation conduction coefficient of the target stock in the transaction process, and calculating a dynamic hidden danger infection entropy value of the target stock; historical position change data and abnormal transaction signals of the target stocks are recorded, the position coupling deviation rate of the target stocks is calculated, the extreme covariant coefficient of the target stocks is calculated, and the strategy fitness between the target stocks is evaluated; and dynamically adjusting the position allocation weight of the target stock, generating a dynamic hedging instruction of the target stock, performing real-time optimization regulation and control on the position of the target stock, and obtaining a combined adjustment strategy of the target stock. The stock staring and investment strategy optimization efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to a method and system for optimizing stock market monitoring and investment strategies in an intelligent investment advisory system, and belongs to the technical field of financial management. Background Art

[0002] In the field of financial investment, stock investment attracts many investors with its high return potential. With the rapid development of information technology and the profound changes in the financial market, the scale of the stock market continues to expand, trading activity continues to increase, the number of various stock products is growing exponentially, and the market environment is becoming more complex and changeable. When participating in stock investment, investors not only need to pay attention to the real-time price fluctuations, trading volume changes, company announcements and other dynamic information of a large number of stocks, but also need to comprehensively consider the macroeconomic situation, industry development trends, company fundamentals and other factors in order to formulate reasonable and effective investment strategies.

[0003] Currently, individual investors face many difficulties when investing in stocks. On the one hand, ordinary investors lack professional financial knowledge and rich investment experience, and find it difficult to accurately interpret complex market data and information. They often feel at a loss when faced with a large number of stock choices and find it difficult to make scientific and reasonable investment decisions. On the other hand, investors have limited time and energy, and are unable to monitor the market in real time, seize investment opportunities brought about by market changes in a timely manner, or avoid potential risks. Moreover, traditional stock investment strategies are often relatively simple and static, and are difficult to adjust in time according to dynamic changes in the market, and cannot adapt to the rapidly changing market environment. Therefore, an efficient method for monitoring the stock market and optimizing investment strategies is needed. Summary of the Invention

[0004] The present invention provides a method and system for optimizing stock market monitoring and investment strategies in an intelligent investment advisory system, the main purpose of which is to improve the efficiency of stock market monitoring and investment strategy optimization.

[0005] To achieve the above objectives, the present invention provides a method for stock market monitoring and investment strategy optimization in a smart investment advisory system, comprising: Acquire multi-dimensional market data of target stocks in the robo-advisory system in real time, analyze the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, collect the sector volatility and liquidity index of the target stocks, and generate a portfolio allocation tier for the target stocks based on the financing scenario classification labels, the sector volatility, and the liquidity index; Calculate the volatility transmission coefficient of the target stock during the trading process, and count the dynamic expenditure costs of the target stock during the holding period. Combine the volatility transmission coefficient and dynamic expenditure costs to calculate the dynamic hidden danger contagion entropy value of the target stock; Recording historical position change data and abnormal trading signals of the target stock, calculating the position coupling deviation rate of the target stock based on the historical position change data, calculating the extreme covariance coefficient of the target stock based on the abnormal trading signal, and evaluating the strategy adaptability between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient; In combination with the field volatility and the dynamic hidden danger infection entropy value, the position allocation weight of the target stock is dynamically adjusted, and according to the strategy fitness and the position allocation weight, a dynamic hedging instruction of the target stock is generated. Based on the dynamic hedging instruction and the allocation tier of the investment combination, the holding of the target stock is optimized and regulated in real time to obtain the portfolio adjustment strategy of the target stock.

[0006] Optionally, parsing the financing scenario classification label corresponding to the target stock based on the multi-dimensional market data includes: Extracting text data from the multidimensional market data to obtain multidimensional market text; Identifying the text part of speech corresponding to the multidimensional market text, and performing text cleaning on the multidimensional market text based on the text part of speech to obtain a target multidimensional market text; Extracting key market text information from the target multidimensional market text, performing knowledge graph construction processing on the key market text information, and obtaining text structured information; Based on the text structured information, a financing scenario classification label corresponding to the target stock is determined.

[0007] Optionally, generating a portfolio allocation tier for the target stock based on the financing scenario classification label, the field volatility, and the liquidity index includes: Performing risk level mapping on the financing scenario classification labels to obtain scenario risk levels; Performing quantile normalization on the volatility of the field to obtain volatility quantile values; Performing segmented normalization processing on the liquidity indicator to obtain a liquidity score; Calculate the comprehensive risk value of the target stock by combining the scenario risk level, the volatility percentile and the liquidity score; Based on the comprehensive risk value, an allocation tier for the target stock's investment portfolio is generated.

[0008] Optionally, the calculating the volatility transmission coefficient of the target stock during the trading process includes: collecting stock trading data of the target stock in real time during the trading process, performing noise reduction processing on the stock trading data, and obtaining optimized stock trading data; extracting the stock transmission characteristics and the stock transaction price from the optimized stock transaction data, performing dimensionality reduction processing on the stock transmission characteristics, and obtaining key stock transmission characteristics; determining a time series relationship between the stock trading price and the key stock transmission characteristic; Based on the time series relationship, the key stock transmission characteristics are sorted to obtain sorted stock transmission characteristics; The volatility transmission coefficient of the target stock during the transaction is calculated by combining the transmission characteristics of the ranked stocks and the stock transaction price.

[0009] Optionally, the calculating of the volatility transmission coefficient of the target stock during the trading process by combining the ranked stock transmission characteristics and the stock trading price includes: Quantifying the transmission characteristics of the ranked stocks to obtain transmission characteristic values; Based on the transmission characteristic value, calculating the characteristic standard deviation corresponding to the transmission characteristic of the sorted stocks; Calculating price differences between adjacent prices in the stock trading price, and calculating price volatility corresponding to the stock trading price based on the price differences; The volatility transmission coefficient of the target stock during the trading process is calculated by combining the characteristic standard deviation and the price volatility.

[0010] Optionally, the step of calculating the dynamic hidden danger contagion entropy value of the target stock by combining the volatility transmission coefficient and the dynamic expenditure fee includes: Normalizing the fluctuation transmission coefficient and the dynamic expenditure cost respectively to obtain a normalized transmission coefficient and a normalized expenditure cost; analyzing the correlation strength between the fluctuation transmission coefficient and the dynamic expenditure cost, and assigning a grey correlation weight between the fluctuation transmission coefficient and the dynamic expenditure cost based on the correlation strength; Combining the grey association weight, the normalized transmission coefficient and the normalized expenditure, the dynamic hidden danger infection entropy value of the target stock is calculated by the following formula: ; Among them, A represents the dynamic hidden danger infection entropy value of the target stock, represents the grey association weight of the ath normalized transmission coefficient and the bth normalized expenditure cost, represents the ath normalized conduction coefficient, represents the bth normalized expenditure, a and b represent the serial numbers corresponding to the normalized conduction coefficient and the normalized expenditure, respectively, and q and w represent the quantities corresponding to the normalized conduction coefficient and the normalized expenditure, respectively.

[0011] Optionally, the calculating the position coupling deviation rate of the target stock based on the historical position change data includes: Smoothing the historical position change data to obtain smoothed position change data; Extracting position-related variables from the smoothed position change data, and calculating the variable Gini index corresponding to the position-related variables; Based on the Gini index of the variable, screening out the variables representing the position among the position-related variables; Calculate the variable information gain corresponding to the position characterizing variable, and combine the variable information gain and the position characterizing variable to calculate the position coupling deviation rate of the target stock using the following formula: ; Among them, E represents the target stock's position coupling deviation rate, represents the variable information gain of the d-th variable in the position variable, Indicates the value of the dth variable in the position variable. It represents the reference value of the dth variable in the variables representing the position, d represents the serial number of the variable representing the position, and p represents the number of variables representing the position.

[0012] Optionally, calculating the extreme covariance coefficient of the target stock based on the abnormal trading signal includes: Determining an extreme event corresponding to the target stock based on the abnormal trading signal; Performing time alignment on the extreme events to obtain aligned extreme events; Counting the event time windows corresponding to the aligned extreme events, and calculating the rate of return of the target stock in the event time windows; Based on the rate of return, calculating the event covariance corresponding to the aligned extreme event; Based on the event covariance, the extreme covariance coefficient of the target stock is calculated.

[0013] Optionally, dynamically adjusting the position allocation weight of the target stock in combination with the field volatility and the dynamic hidden danger contagion entropy value includes: The position allocation weight of the target stock is dynamically adjusted using the following formula: ; Among them, G represents the position allocation weight of the target stock, M represents the volatility of the field, and N represents the dynamic hidden danger infection entropy value.

[0014] In order to solve the above problems, the present invention further provides a stock market monitoring and investment strategy optimization system in a smart investment advisory system, the system comprising: A module for generating an investment portfolio allocation tier is configured to obtain multi-dimensional market data of target stocks in the robo-advisory system in real time, analyze the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, collect the sector volatility and liquidity index of the target stocks, and generate an investment portfolio allocation tier for the target stocks based on the financing scenario classification labels, the sector volatility, and the liquidity index; A dynamic hidden danger transmission entropy value calculation module is used to calculate the volatility transmission coefficient of the target stock during the trading process, and to count the dynamic expenditure costs of the target stock during the holding period. The dynamic hidden danger transmission entropy value of the target stock is calculated by combining the volatility transmission coefficient and the dynamic expenditure costs; a strategy fitness evaluation module, configured to record the historical position change data and abnormal trading signals of the target stocks, calculate the position coupling deviation rate of the target stocks based on the historical position change data, calculate the extreme covariance coefficient of the target stocks based on the abnormal trading signals, and evaluate the strategy fitness between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient; A position optimization and control module is used to dynamically adjust the position allocation weight of the target stock in combination with the field volatility and the dynamic hidden danger infection entropy value, generate dynamic hedging instructions for the target stock according to the strategy fitness and the position allocation weight, and optimize and control the position of the target stock in real time based on the dynamic hedging instructions and the allocation tiers of the investment combination to obtain the combination adjustment strategy of the target stock.

[0015] Compared to the problems described in the background art, the present invention, by analyzing the financing scenario classification labels corresponding to the target stock based on the multi-dimensional market data, allows investors to clearly understand the motivations and actual conditions of corporate financing and gain early insight into potential risks and growth opportunities. Furthermore, by calculating the volatility transmission coefficient of the target stock during the trading process, the present invention can use this volatility transmission coefficient to understand the transmission pattern of stock price fluctuations within the market system. By statistically analyzing the dynamic expenditure costs of the target stock during the holding period, it can clearly understand the dynamic changes in the actual cost of holding the stock, laying an important foundation for the subsequent calculation of the dynamic hidden danger contagion entropy value. By calculating the holding coupling deviation rate of the target stock based on the historical holding change data, the present invention can understand the degree of deviation between the dynamic changes in the target stock's holding structure and the normal market state, helping investors to promptly detect holding anomalies. Furthermore, by combining the field volatility and the dynamic hidden danger contagion entropy value to dynamically adjust the position allocation weight of the target stock, the present invention can effectively reflect the impact of market fluctuations and potential risks on the target stock, providing a scientific basis for the subsequent generation of dynamic hedging instructions and optimization of holdings. Therefore, the stock market monitoring and investment strategy optimization method and system in the intelligent investment advisory system provided by the embodiment of the present invention can improve the efficiency of stock market monitoring and investment strategy optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a method for stock market monitoring and investment strategy optimization in a smart investment advisory system provided by one embodiment of the present invention; Figure 2 A schematic diagram of a module for implementing a method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to an embodiment of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] The present embodiment provides a method for stock market monitoring and investment strategy optimization in a robo-advisory system. The execution entity of this method includes, but is not limited to, at least one of a server, a terminal, or other electronic device capable of executing the method provided in the present embodiment. In other words, the method can be executed by software or hardware installed on a terminal or server device. The server may include, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0020] Example 1: Reference Figure 1 FIG. 1 is a flow chart of a method for optimizing stock market trends and investment strategies in a smart investment advisory system according to an embodiment of the present invention. In this embodiment, the method for optimizing stock market trends and investment strategies in a smart investment advisory system includes: S1. Acquire multi-dimensional market data of target stocks in the smart investment advisory system in real time, analyze the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, collect the field volatility and liquidity indicators of the target stocks, and generate the investment portfolio allocation tiers of the target stocks according to the financing scenario classification labels, the field volatility and the liquidity indicators.

[0021] The present invention analyzes the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, so that investors can clearly understand the motivations and actual conditions of corporate financing and gain insight into potential risks and growth opportunities in advance. It should be explained that the intelligent investment advisory system is a digital platform that uses artificial intelligence technology to provide investment advice to investors. The target stocks are specific stocks selected by investors in the intelligent investment advisory system as the objects of investment analysis and decision-making. The multi-dimensional market data are real-time or historical data of the target stocks in the intelligent investment advisory system covering multiple dimensions such as transactions, finances, and industry dynamics. The financing scenario classification labels are specific identifiers corresponding to the target stocks for identifying their financing models, motivations and related situational categories. Furthermore, the multi-dimensional market data of the target stocks in the intelligent investment advisory system can be obtained in real time through a data crawler program.

[0022] Specifically, parsing the financing scenario classification label corresponding to the target stock based on the multi-dimensional market data includes: Extracting text data from the multidimensional market data to obtain multidimensional market text; Identifying the text part of speech corresponding to the multidimensional market text, and performing text cleaning on the multidimensional market text based on the text part of speech to obtain a target multidimensional market text; Extracting key market text information from the target multidimensional market text, performing knowledge graph construction processing on the key market text information, and obtaining text structured information; Based on the text structured information, a financing scenario classification label corresponding to the target stock is determined.

[0023] It should be explained that the multidimensional market text is the part of the multidimensional market data presented in text form, such as company announcements, news reports, research reports and other text materials; the text part of speech is the grammatical property of each word corresponding to the multidimensional market text, such as noun, verb, adjective, etc., which is used to analyze the language structure and semantics of the text; the target multidimensional market text is the part of the text that meets specific analysis requirements after the multidimensional market text is screened, classified or labeled based on the text part of speech; the market key text information is the text content in the target multidimensional market text that can reflect the key market characteristics, trends, events or influencing factors and other important values; the text structured information is the result of constructing a knowledge graph for the market key text information, which displays the information such as entities, relationships and attributes in the text in a graphical and structured manner, so as to facilitate in-depth understanding and analysis of market information.

[0024] Furthermore, text data extraction can be performed on the multidimensional market data using regular expression matching and web crawlers in natural language processing technology in combination with information extraction algorithms to obtain multidimensional market text. Part-of-speech tagging tools such as NLTK (Natural Language Toolkit) and Stanford Core NLP can be used to identify the text parts of speech corresponding to the multidimensional market text. Based on the text parts of speech, the multidimensional market text can be cleaned using rules and algorithms such as stop word filtering, part-of-speech screening, and lexical and syntactic analysis to obtain target multidimensional market text. Key market text information can be extracted from the target multidimensional market text using keyword extraction algorithms such as TF-IDF and TextRank. A knowledge graph can be constructed for the key market text information using graph databases such as Neo4j and JanusGraph in combination with entity recognition and relationship extraction technologies to obtain text structured information. Based on the text structured information, a financing scenario classification label corresponding to the target stock can be determined. For example, when the text structured information indicates that the company obtains funds by issuing bonds and uses them to expand production scale, its financing scenario classification label can be determined as debt financing for capacity expansion.

[0025] The present invention generates a portfolio allocation tier for the target stock based on the financing scenario classification label, the sector volatility, and the liquidity index, thereby making hierarchical and sequential arrangements for the allocation of investment funds to the target stock. It should be explained that the sector volatility and liquidity index are important quantitative reflections of the target stock's market environment and its own trading characteristics. The former reflects the severity and uncertainty of price fluctuations in the sector in which the target stock resides, while the latter measures the target stock's ability to quickly trade at a reasonable price without significantly impacting the market price during the trading process. The portfolio allocation tier is a specific strategic arrangement for the hierarchical and sequential allocation of investment funds based on a comprehensive consideration of multiple factors, including the target stock's own characteristics, financing scenario, and market environment. It is used to optimize the investment portfolio and balance risk and return. Furthermore, the collection of the target stock's sector volatility and liquidity index can be achieved through professional interfaces of financial data providers (such as Wind and Bloomberg data interfaces).

[0026] In detail, generating the target stock's investment portfolio allocation tiers based on the financing scenario classification label, the field volatility, and the liquidity index includes: Performing risk level mapping on the financing scenario classification labels to obtain scenario risk levels; Performing quantile normalization on the volatility of the field to obtain volatility quantile values; Performing segmented normalization processing on the liquidity indicator to obtain a liquidity score; Calculate the comprehensive risk value of the target stock by combining the scenario risk level, the volatility percentile and the liquidity score; Based on the comprehensive risk value, an allocation tier for the target stock's investment portfolio is generated.

[0027] It should be explained that the scenario risk level is a quantitative level identifier obtained after the financing scenario classification label is mapped to the risk level, reflecting the potential risk level of different financing scenarios; the volatility percentile is a quantitative value obtained after the field volatility is normalized by percentile, which can intuitively show the position of the field volatility in the overall distribution; the liquidity score is a quantitative score generated after the liquidity indicator is segmented and normalized, used to measure the liquidity level of the target stock; the comprehensive risk value is a comprehensive quantitative risk value for the target stock calculated by combining the scenario risk level, the volatility percentile and the liquidity score, which comprehensively considers the financing scenario risk, field volatility risk and liquidity risk. This value is used to evaluate the overall risk status of the target stock under multi-dimensional risk factors.

[0028] Furthermore, the financing scenario classification labels can be mapped to risk levels using a pre-set risk level mapping rule library to obtain scenario risk levels; the volatility of the field can be quantile-normalized using professional statistical analysis software (such as the statistical analysis library in R language and Python) to obtain volatility quantile values; the liquidity indicator can be segmented and normalized using a custom segmentation interval and normalization calculation formula to obtain a liquidity score; in combination with the scenario risk level, the volatility quantile value and the liquidity score, the weight of each factor is determined using the analytic hierarchy process (AHP) or entropy weight method, and the comprehensive risk value of the target stock is calculated in a weighted summation manner; based on the comprehensive risk value, an allocation tier for the target stock's investment portfolio is generated, such as sorting the comprehensive risk values ​​from low to high, prioritizing the allocation of funds to stocks with low comprehensive risk values, and determining the fund allocation ratio and investment order of different stocks in order of increasing risk.

[0029] S2. Calculate the volatility transmission coefficient of the target stock during the transaction process, and count the dynamic expenditure costs of the target stock during the holding period. Combine the volatility transmission coefficient and the dynamic expenditure costs to calculate the dynamic hidden danger contagion entropy value of the target stock.

[0030] The present invention calculates the volatility transmission coefficient of the target stock during the transaction process, and can grasp the transmission law of stock price fluctuations in the market system through the volatility transmission coefficient. By counting the dynamic expenditure costs of the target stock during the holding period, it can clearly understand the dynamic changes in the actual cost of holding the stock, and lay an important basis for the subsequent calculation of the dynamic hidden danger infection entropy value. It should be explained that the volatility transmission coefficient is a quantitative indicator to measure the impact of the target stock price fluctuation on other related stocks, market sectors and even the entire financial market, reflecting the speed, intensity and direction of the wave propagation; the dynamic expenditure costs cover the total costs incurred by investors during the holding period due to various transaction fees, taxes and capital costs that change with the market environment. Furthermore, the dynamic expenditure costs of the target stock during the holding period can be obtained by using an automated data processing program to count the investor's capital flow records.

[0031] Specifically, the calculation of the volatility transmission coefficient of the target stock during the trading process includes: collecting stock trading data of the target stock in real time during the trading process, performing noise reduction processing on the stock trading data, and obtaining optimized stock trading data; extracting the stock transmission characteristics and the stock transaction price from the optimized stock transaction data, performing dimensionality reduction processing on the stock transmission characteristics, and obtaining key stock transmission characteristics; determining a time series relationship between the stock trading price and the key stock transmission characteristic; Based on the time series relationship, the key stock transmission characteristics are sorted to obtain sorted stock transmission characteristics; The volatility transmission coefficient of the target stock during the transaction is calculated by combining the transmission characteristics of the ranked stocks and the stock transaction price.

[0032] It should be explained that the stock trading data are various types of relevant data of the target stock during the trading process, such as transaction price, transaction volume, transaction time, buy and sell order information, etc.; the optimized stock trading data are the stock trading data that have been processed by noise reduction, cleaning, etc. to remove outliers, noise interference, etc. to improve data quality and availability; the stock transmission characteristics and the stock trading prices are respectively the relevant attribute data (such as turnover rate, capital flow, etc.) and the stock transaction price numerical information in the optimized stock trading data that can reflect the stock volatility transmission situation; the key stock transmission characteristics are the characteristics that play an important role in stock volatility transmission that are screened out by dimensionality reduction processing and removing redundant information of the stock transmission characteristics; the time series relationship is the correlation between the stock trading price and the key stock transmission characteristics in the time sequence and the mutual influence relationship that changes over time; the sorted stock transmission characteristics are the feature set obtained by sorting the key stock transmission characteristics based on the time series relationship.

[0033] Furthermore, the stock trading data of the target stock during the trading process can be collected in real time through the above-mentioned Wind and Bloomberg data interfaces, and the stock trading data can be subjected to noise reduction processing through algorithms such as moving average filtering and wavelet transform to obtain optimized stock trading data; the stock transmission characteristics and the stock trading prices in the optimized stock trading data can be extracted through data mining techniques (such as association rule mining, feature selection algorithms, etc.); the stock transmission characteristics can be subjected to dimensionality reduction processing through dimensionality reduction methods such as principal component analysis (PCA) and factor analysis to obtain key stock transmission characteristics; the time series relationship between the stock trading price and the key stock transmission characteristics can be determined through time series analysis methods (such as autocorrelation function and cross-correlation function calculation); based on the time series relationship, the key stock transmission characteristics can be sorted according to rules such as chronological order or correlation strength to obtain sorted stock transmission characteristics.

[0034] Furthermore, as an optional embodiment of the present invention, the calculation of the volatility transmission coefficient of the target stock during the trading process by combining the ranked stock transmission characteristics and the stock trading price includes: Quantifying the transmission characteristics of the ranked stocks to obtain transmission characteristic values; Based on the transmission characteristic value, calculating the characteristic standard deviation corresponding to the transmission characteristic of the sorted stocks; Calculating price differences between adjacent prices in the stock trading price, and calculating price volatility corresponding to the stock trading price based on the price differences; The volatility transmission coefficient of the target stock during the trading process is calculated by combining the characteristic standard deviation and the price volatility.

[0035] It should be explained that the conduction characteristic value is the specific numerical value obtained after the conduction characteristic of the sorted stock is quantified; the characteristic standard deviation is a measure of the degree of dispersion of the data sequence corresponding to the conduction characteristic of the sorted stock; the price difference value is the difference obtained by subtracting two adjacent prices in the stock trading price, and the price volatility is a measure of the fluctuation amplitude of the data sequence corresponding to the stock trading price.

[0036] Furthermore, the transmission characteristics of the ranked stocks can be quantified by a coding algorithm to obtain a transmission characteristic value, such as a one-hot encoding algorithm; based on the transmission characteristic value, the characteristic standard deviation corresponding to the transmission characteristics of the ranked stocks can be calculated by the standard deviation calculation formula in statistics; based on the price difference value, the price volatility corresponding to the stock trading price can be calculated by the annualized volatility formula or the moving average standard deviation and other methods; the calculation formula for the volatility transmission coefficient of the target stock during the transaction is: the characteristic standard deviation × the price volatility × the benchmark volatility, where the benchmark volatility is a coefficient related to the overall market volatility or the average industry volatility.

[0037] The present invention calculates the dynamic hidden danger contagion entropy value of the target stock by combining the volatility transmission coefficient and the dynamic expenditure cost, which can comprehensively consider the impact of fluctuations in stock transactions and actual cost consumption on potential risks, and provide investors with a comprehensive and quantitative risk assessment indicator. It should be explained that the dynamic hidden danger contagion entropy value is a key indicator for measuring the investment risk of the target stock during the transaction process.

[0038] Specifically, the calculation of the dynamic hidden danger contagion entropy value of the target stock by combining the volatility transmission coefficient and the dynamic expenditure fee includes: Normalizing the fluctuation transmission coefficient and the dynamic expenditure cost respectively to obtain a normalized transmission coefficient and a normalized expenditure cost; analyzing the correlation strength between the fluctuation transmission coefficient and the dynamic expenditure cost, and assigning a grey correlation weight between the fluctuation transmission coefficient and the dynamic expenditure cost based on the correlation strength; Combining the grey association weight, the normalized transmission coefficient and the normalized expenditure, the dynamic hidden danger infection entropy value of the target stock is calculated by the following formula: ; Among them, A represents the dynamic hidden danger infection entropy value of the target stock, represents the grey association weight of the ath normalized transmission coefficient and the bth normalized expenditure cost, represents the ath normalized conduction coefficient, represents the bth normalized expenditure, a and b represent the serial numbers corresponding to the normalized conduction coefficient and the normalized expenditure, respectively, and q and w represent the quantities corresponding to the normalized conduction coefficient and the normalized expenditure, respectively.

[0039] It should be explained that the normalized transmission coefficient and the normalized expenditure cost are the values ​​obtained after the dimensional differences between the volatility transmission coefficient and the dynamic expenditure cost are eliminated. The correlation strength indicates the degree of correlation between the volatility transmission coefficient and the dynamic expenditure cost, and is used to measure the impact of the change of one factor on another factor. The grey correlation weight is a value determined based on the correlation strength between the volatility transmission coefficient and the dynamic expenditure cost, reflecting the relative importance of their respective impacts on the potential risks of the target stock. It is used to reasonably measure the role of the two factors when calculating the dynamic hidden danger contagion entropy value.

[0040] Furthermore, the fluctuation transmission coefficient and dynamic expenditure cost can be normalized respectively by the Z-score normalization method and the Min-Max normalization method to obtain the normalized transmission coefficient and the normalized expenditure cost; the correlation strength between the fluctuation transmission coefficient and the dynamic expenditure cost can be analyzed by the grey correlation analysis method, and based on the correlation strength, the grey correlation weight between the fluctuation transmission coefficient and the dynamic expenditure cost is allocated according to the degree of correlation.

[0041] S3. Record the historical position change data and abnormal trading signals of the target stocks, calculate the position coupling deviation rate of the target stocks based on the historical position change data, calculate the extreme covariance coefficient of the target stocks based on the abnormal trading signals, and evaluate the strategy adaptability between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient.

[0042] By calculating the target stock's position coupling deviation rate based on the historical position change data, the present invention can provide insight into the degree of deviation between the dynamic changes in the target stock's holding structure and the normal market state, assisting investors in promptly identifying abnormal holdings. It should be explained that the historical position change data is a data set consisting of relevant information such as the number of investors holding the target stock, time points, and increase / decrease operations during past trading periods, reflecting the dynamic changes in the target stock's holdings over time. Abnormal trading signals are trading characteristics or indicator changes that significantly deviate from normal trading patterns, market expectations, or historical data performance during the trading process of the target stock, indicating the possibility of special market conditions or potential risks. The position coupling deviation rate is a quantitative indicator of the deviation between the current holding state and the expected or normal holding pattern for the target stock, calculated using a specific calculation method based on the historical position change data. It can help investors determine the rationality of their holding structure and potential risks. Furthermore, the recording of the target stock's historical position change data and abnormal trading signals can be implemented through financial trading software or platforms (such as securities companies' trading terminals, Wind Information, etc.).

[0043] Specifically, the calculating the target stock's position coupling deviation rate based on the historical position change data includes: Smoothing the historical position change data to obtain smoothed position change data; Extracting position-related variables from the smoothed position change data, and calculating the variable Gini index corresponding to the position-related variables; Based on the Gini index of the variable, screening out the variables representing the position among the position-related variables; Calculate the variable information gain corresponding to the position characterizing variable, and combine the variable information gain and the position characterizing variable to calculate the position coupling deviation rate of the target stock using the following formula: ; Among them, E represents the target stock's position coupling deviation rate, represents the variable information gain of the d-th variable in the position variable, Indicates the value of the dth variable in the position variable. It represents the reference value of the dth variable in the variables representing the position, d represents the serial number of the variable representing the position, and p represents the number of variables representing the position.

[0044] It should be explained that the smoothed position change data is the data after the historical position change data has been smoothed; the position-related variables are specific data indicators in the smoothed position change data; the variable Gini index is the quantitative value of the importance corresponding to the position-related variables; the variable representing the position is the key variable screened out from the position-related variables based on the variable Gini index; the variable information gain is the quantitative value of the contribution degree corresponding to the variable representing the position.

[0045] Furthermore, the historical position change data can be smoothed by an exponential smoothing algorithm to obtain smoothed position change data; the position-related variables in the smoothed position change data can be extracted by an association rule mining method; the variable Gini index corresponding to the position-related variable can be calculated by a Gini index calculation formula; the variable Gini index is compared with a set Gini index threshold, and variables above the threshold are used as screening results to obtain the characterizing position variables in the position-related variables; the variable information gain corresponding to the characterizing position variable can be calculated by a Shannon entropy algorithm; the reference value of the characterizing position variable can be obtained by analyzing the typical value range, mean, median and other statistics of the variable in historical data.

[0046] The present invention calculates the extreme covariance coefficient of the target stock based on the abnormal trading signal. The extreme covariance coefficient can reflect the coordinated changes of the target stock under extreme market conditions, laying an important basis for the subsequent evaluation of the strategy adaptability between the target stocks. It should be explained that the extreme covariance coefficient represents a quantitative indicator of the degree of correlation and coordinated changes between the price changes or returns of the target stock and other related assets (such as market indices, stocks in the same industry, etc.) or the overall market environment under extreme market conditions (such as a sharp drop or surge in the stock market).

[0047] Specifically, the calculating of the extreme covariance coefficient of the target stock based on the abnormal trading signal includes: Determining an extreme event corresponding to the target stock based on the abnormal trading signal; Performing time alignment on the extreme events to obtain aligned extreme events; Counting the event time windows corresponding to the aligned extreme events, and calculating the rate of return of the target stock in the event time window; Based on the rate of return, calculating the event covariance corresponding to the aligned extreme event; Based on the event covariance, the extreme covariance coefficient of the target stock is calculated.

[0048] It should be explained that the extreme events are the abnormal situations that are obviously different from the normal trading status, such as sharp fluctuations in stock prices in a short period of time, sudden changes in trading volume, etc., which may cause major market impacts or reflect special investment behaviors during the stock trading process of the target stocks. The aligned extreme events are to align the extreme events in time so that different extreme events of the target stocks are in comparable positions on the time axis, ensuring that the time points of the relevant data of each extreme event (such as trading data) can be accurately matched, so as to facilitate subsequent analysis and research on extreme events in a processing state. The event time window is the time window set around the aligned extreme events for in-depth analysis of the aligned extreme events. The specific time range of the extreme event period is used to collect target stocks and related data within this range for targeted research; the yield rate is the profit or loss generated by factors such as the target stock's stock price fluctuations and dividends within the event time window, and the ratio value reflecting the investment return of the period is obtained in a certain calculation method; the event covariance is a statistical indicator corresponding to the aligned extreme event, which is used to quantify the degree of correlation between the yield rates of the target stock and other related assets (such as reference stocks, market indices, etc.) within the time interval determined by the aligned extreme event, reflecting their coordinated change relationship during the extreme event.

[0049] Furthermore, based on the abnormal trading signal, the extreme event corresponding to the target stock is determined according to a pre-set abnormal event determination rule (formulated in combination with market experience and financial theory); the extreme event can be time-aligned by comparing the time of occurrence of the extreme event with a unified time benchmark (such as the opening time of the trading day) to obtain an aligned extreme event; the event time window corresponding to the aligned extreme event can be counted by sorting out the start and end time points of the aligned extreme event, and the return rate of the target stock in the event time window can be calculated by calculating the difference between the closing price and the opening price of the stock in the event time window and dividing it by the opening price; based on the return rate, the event covariance corresponding to the aligned extreme event can be calculated by the covariance calculation formula (taking into account the combined relationship of the returns at different time points); based on the event covariance, the extreme covariance coefficient of the target stock is calculated by a specific coefficient calculation model (taking into account the event covariance and other relevant factors).

[0050] The present invention evaluates the strategy fitness between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient. The strategy fitness can be used to understand the degree of investment strategy adaptation between the target stocks, and to judge whether the current investment strategy can effectively cope with the changes in the target stock position structure and extreme market conditions, thereby laying an important foundation for the subsequent generation and processing of dynamic hedging instructions for the target stocks. It should be explained that the strategy fitness is the degree of investment strategy adaptation between the target stocks. Further, in combination with the position coupling deviation rate and the extreme covariance coefficient, the strategy fitness between the target stocks is evaluated. For example, when the position coupling deviation rate is low and the extreme covariance coefficient is small, it indicates that the investment strategy is highly consistent with the target stock's position structure and extreme market volatility. For example, during a certain period of time, the position coupling deviation rate of stock A and stock B is only 5%, and the extreme covariance coefficient is 0.1, indicating that the current investment strategy can better adapt to the position changes and extreme situations of these two stocks and can continue to be used. On the contrary, if the position coupling deviation rate is high and the extreme covariance coefficient is large, such as the position coupling deviation rate of stock C and stock D reaching 30%, and the extreme covariance coefficient is 0.8, it means that the investment strategy may not be effectively adapted and the strategy needs to be adjusted in time to reduce risks.

[0051] S4. Dynamically adjust the position allocation weight of the target stock in combination with the field volatility and the dynamic hidden danger infection entropy value; generate a dynamic hedging instruction for the target stock based on the strategy fitness and the position allocation weight; and optimize and regulate the holdings of the target stock in real time based on the dynamic hedging instruction and the allocation tiers of the investment portfolio to obtain a portfolio adjustment strategy for the target stock.

[0052] The present invention dynamically adjusts the position allocation weight of the target stock by combining the field volatility and the dynamic hidden danger infection entropy value, which can effectively reflect the impact of market fluctuations and potential risks on the target stock, and provide a scientific basis for the subsequent generation of dynamic hedging instructions and optimization of positions. It should be explained that the position allocation weight is the allocation ratio of the target stock in the investment portfolio.

[0053] Specifically, the dynamically adjusting the position allocation weight of the target stock by combining the field volatility and the dynamic hidden danger contagion entropy value includes: The position allocation weight of the target stock is dynamically adjusted using the following formula: ; Among them, G represents the position allocation weight of the target stock, M represents the volatility of the field, and N represents the dynamic hidden danger infection entropy value.

[0054] The present invention generates dynamic hedging instructions for the target stock based on the strategy fitness and the position allocation weight, which allows investors to accurately formulate hedging operation plans based on the degree of fit between the investment strategy and the stock characteristics and the capital layout, effectively reduce investment risks, and improve the stability and return level of the investment portfolio. It should be explained that the dynamic hedging instructions are generated by the target stock based on its strategy fitness and position allocation weight, containing detailed information such as the type, quantity, buying and selling direction, and trading timing of the hedged assets, so as to cope with market fluctuations and potential risks, adjust the position structure in real time, and ensure the stability and return level of the investment portfolio. Furthermore, the generation step of the dynamic hedging instructions for the target stock is: determine the correlation between the strategy fitness and the position weight. If the strategy fitness is high, it means that the investment strategy is well matched with the target stock, and the position weight based on the strategy fitness can be appropriately increased. The hedging intensity of the position weight is determined; if the strategy fitness is low, the hedging intensity is reduced. For example, when the strategy fitness score is above 80 points (out of 100), each 1% of the position weight corresponds to 1.2 units of hedged assets; when the strategy fitness score is between 60 and 80 points, each 1% of the position weight corresponds to 1 unit of hedged assets; when the strategy fitness score is below 60 points, each 1% of the position weight corresponds to 0.8 units of hedged assets. The required amount of hedged assets is calculated based on the association relationship and the current position weight allocation. Assuming the current position weight is 30% and the strategy fitness score is 75 points, the required amount of hedged assets = 30% × 1 × total portfolio size. Based on the calculation results, a dynamic hedging instruction is generated, containing detailed information such as the type, quantity, and transaction direction of the hedged assets; for example, the instruction is to buy stock index futures contracts worth X yuan for hedging.

[0055] The present invention optimizes and controls the holdings of the target stock in real time based on the dynamic hedging instructions and the allocation tiers of the investment portfolio, optimizes and controls the holdings of the target stock accurately and efficiently in real time, and then formulates a portfolio adjustment strategy for the target stock that meets actual needs and is practical based on a comprehensive consideration of market dynamics, investment risks and expected returns. Furthermore, based on the dynamic hedging instructions and the allocation tiers of the investment portfolio, the holdings of the target stock are optimized and controlled in real time to obtain a portfolio adjustment strategy for the target stock. For example, according to the transaction details of the hedging tools specified in the dynamic hedging instructions, the buying and selling operations are executed in sequence according to the order and capital ratio set in the allocation tiers of the investment portfolio, and the target stock holding quantity is dynamically adjusted, thereby forming a target stock portfolio adjustment strategy that includes operation records of each stage, position adjustment range and final position configuration.

[0056] Compared to the problems described in the background art, the present invention, by analyzing the financing scenario classification labels corresponding to the target stock based on the multi-dimensional market data, allows investors to clearly understand the motivations and actual conditions of corporate financing and gain early insight into potential risks and growth opportunities. Furthermore, by calculating the volatility transmission coefficient of the target stock during the trading process, the present invention can use this volatility transmission coefficient to understand the transmission pattern of stock price fluctuations within the market system. By statistically analyzing the dynamic expenditure costs of the target stock during the holding period, it can clearly understand the dynamic changes in the actual cost of holding the stock, laying an important foundation for the subsequent calculation of the dynamic hidden danger contagion entropy value. By calculating the holding coupling deviation rate of the target stock based on the historical holding change data, the present invention can understand the degree of deviation between the dynamic changes in the target stock's holding structure and the normal market state, helping investors to promptly detect holding anomalies. Furthermore, by combining the field volatility and the dynamic hidden danger contagion entropy value to dynamically adjust the position allocation weight of the target stock, the present invention can effectively reflect the impact of market fluctuations and potential risks on the target stock, providing a scientific basis for the subsequent generation of dynamic hedging instructions and optimization of holdings. Therefore, the stock market monitoring and investment strategy optimization method and system in the intelligent investment advisory system provided by the embodiment of the present invention can improve the efficiency of stock market monitoring and investment strategy optimization.

[0057] Example 2: like Figure 2 The figure shows a functional module diagram of a stock monitoring and investment strategy optimization system in an intelligent investment advisory system of the present invention.

[0058] The stock market monitoring and investment strategy optimization system 200 in a robo-advisory system described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the system can include a portfolio allocation tier generation module 201, a dynamic hidden danger infection entropy calculation module 202, a strategy fitness assessment module 203, and a position optimization and control module 204. A module, also referred to as a unit, refers to a series of computer program segments that can be executed by an electronic device processor and perform a fixed function, and is stored in the electronic device's memory.

[0059] In the embodiment of the present invention, the functions of each module / unit are as follows: The investment portfolio allocation tier generation module 201 is configured to obtain multi-dimensional market data of target stocks in the robo-advisory system in real time, analyze the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, collect the sector volatility and liquidity index of the target stocks, and generate investment portfolio allocation tiers for the target stocks based on the financing scenario classification labels, the sector volatility, and the liquidity index; The dynamic hidden danger transmission entropy value calculation module 202 is used to calculate the volatility transmission coefficient of the target stock during the trading process, and to count the dynamic expenditure costs of the target stock during the holding period. The dynamic hidden danger transmission entropy value of the target stock is calculated by combining the volatility transmission coefficient and the dynamic expenditure costs. The strategy fitness evaluation module 203 is used to record the historical position change data and abnormal trading signals of the target stocks, calculate the position coupling deviation rate of the target stocks based on the historical position change data, calculate the extreme covariance coefficient of the target stocks based on the abnormal trading signals, and evaluate the strategy fitness between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient; The position optimization and control module 204 is used to dynamically adjust the position allocation weight of the target stock in combination with the field volatility and the dynamic hidden danger infection entropy value, generate dynamic hedging instructions for the target stock according to the strategy fitness and the position allocation weight, and optimize and control the position of the target stock in real time based on the dynamic hedging instructions and the allocation tiers of the investment combination to obtain the combination adjustment strategy of the target stock.

[0060] In detail, the modules in the stock market monitoring and investment strategy optimization system 200 in the smart investment advisory system according to the embodiment of the present invention are used in the same manner as above. Figure 1 The stock monitoring and investment strategy optimization methods in the smart investment advisory system described in the article use the same technical means and can produce the same technical effects, so they will not be repeated here.

[0061] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for stock market monitoring and investment strategy optimization in a smart investment advisory system, characterized in that: The method comprises: Acquire multi-dimensional market data of target stocks in the robo-advisory system in real time, analyze the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, collect the sector volatility and liquidity index of the target stocks, and generate a portfolio allocation tier for the target stocks based on the financing scenario classification labels, the sector volatility, and the liquidity index; Calculate the volatility transmission coefficient of the target stock during the trading process, and count the dynamic expenditure costs of the target stock during the holding period. Combine the volatility transmission coefficient and dynamic expenditure costs to calculate the dynamic hidden danger contagion entropy value of the target stock; Recording historical position change data and abnormal trading signals of the target stock, calculating the position coupling deviation rate of the target stock based on the historical position change data, calculating the extreme covariance coefficient of the target stock based on the abnormal trading signal, and evaluating the strategy adaptability between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient; In combination with the field volatility and the dynamic hidden danger infection entropy value, the position allocation weight of the target stock is dynamically adjusted, and according to the strategy fitness and the position allocation weight, a dynamic hedging instruction of the target stock is generated. Based on the dynamic hedging instruction and the allocation tier of the investment combination, the holding of the target stock is optimized and regulated in real time to obtain the portfolio adjustment strategy of the target stock.

2. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, characterized in that: The step of analyzing the financing scenario classification label corresponding to the target stock based on the multi-dimensional market data includes: Extracting text data from the multidimensional market data to obtain multidimensional market text; Identifying the text part of speech corresponding to the multidimensional market text, and performing text cleaning on the multidimensional market text based on the text part of speech to obtain a target multidimensional market text; Extracting key market text information from the target multidimensional market text, performing knowledge graph construction processing on the key market text information, and obtaining text structured information; Based on the text structured information, a financing scenario classification label corresponding to the target stock is determined.

3. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, characterized in that: Generating the target stock's investment portfolio allocation tiers based on the financing scenario classification label, the field volatility, and the liquidity indicator includes: Performing risk level mapping on the financing scenario classification labels to obtain scenario risk levels; Performing quantile normalization on the volatility of the field to obtain volatility quantile values; Performing segmented normalization processing on the liquidity indicator to obtain a liquidity score; Calculate the comprehensive risk value of the target stock by combining the scenario risk level, the volatility percentile and the liquidity score; Based on the comprehensive risk value, an allocation tier for the target stock's investment portfolio is generated.

4. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, wherein: The calculating of the volatility transmission coefficient of the target stock during the trading process includes: collecting stock trading data of the target stock in real time during the trading process, performing noise reduction processing on the stock trading data, and obtaining optimized stock trading data; extracting the stock transmission characteristics and the stock transaction price from the optimized stock transaction data, performing dimensionality reduction processing on the stock transmission characteristics, and obtaining key stock transmission characteristics; determining a time series relationship between the stock trading price and the key stock transmission characteristic; Based on the time series relationship, the key stock transmission characteristics are sorted to obtain sorted stock transmission characteristics; The volatility transmission coefficient of the target stock during the transaction is calculated by combining the transmission characteristics of the ranked stocks and the stock transaction price.

5. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 4, characterized in that: The step of calculating the volatility transmission coefficient of the target stock during the trading process by combining the ranked stock transmission characteristics and the stock trading price includes: Quantifying the transmission characteristics of the ranked stocks to obtain transmission characteristic values; Based on the transmission characteristic value, calculating the characteristic standard deviation corresponding to the transmission characteristic of the sorted stocks; Calculating price differences between adjacent prices in the stock trading price, and calculating price volatility corresponding to the stock trading price based on the price differences; The volatility transmission coefficient of the target stock during the trading process is calculated by combining the characteristic standard deviation and the price volatility.

6. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, wherein: The step of calculating the dynamic hidden danger contagion entropy value of the target stock by combining the volatility transmission coefficient and the dynamic expenditure fee includes: Normalizing the fluctuation transmission coefficient and the dynamic expenditure cost respectively to obtain a normalized transmission coefficient and a normalized expenditure cost; analyzing the correlation strength between the fluctuation transmission coefficient and the dynamic expenditure cost, and assigning a grey correlation weight between the fluctuation transmission coefficient and the dynamic expenditure cost based on the correlation strength; Combining the grey association weight, the normalized transmission coefficient and the normalized expenditure, the dynamic hidden danger infection entropy value of the target stock is calculated by the following formula: ； Among them, A represents the dynamic hidden danger infection entropy value of the target stock, represents the grey association weight of the ath normalized transmission coefficient and the bth normalized expenditure cost, represents the ath normalized conduction coefficient, represents the bth normalized expenditure, a and b represent the serial numbers corresponding to the normalized conduction coefficient and the normalized expenditure, respectively, and q and w represent the quantities corresponding to the normalized conduction coefficient and the normalized expenditure, respectively.

7. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, wherein: The calculating the target stock's position coupling deviation rate based on the historical position change data includes: Smoothing the historical position change data to obtain smoothed position change data; Extracting position-related variables from the smoothed position change data, and calculating the variable Gini index corresponding to the position-related variables; Based on the Gini index of the variable, screening out the variables representing the holdings among the holding-related variables; Calculate the variable information gain corresponding to the position characterizing variable, and combine the variable information gain and the position characterizing variable to calculate the position coupling deviation rate of the target stock using the following formula: ； Among them, E represents the target stock's position coupling deviation rate, represents the variable information gain of the d-th variable in the position variable, Indicates the value of the dth variable in the position variable. It represents the reference value of the dth variable in the variables representing the position, d represents the serial number of the variable representing the position, and p represents the number of variables representing the position.

8. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, wherein: The calculating the extreme covariance coefficient of the target stock based on the abnormal trading signal includes: Determining an extreme event corresponding to the target stock based on the abnormal trading signal; Performing time alignment on the extreme events to obtain aligned extreme events; Counting the event time windows corresponding to the aligned extreme events, and calculating the rate of return of the target stock in the event time windows; Based on the rate of return, calculating the event covariance corresponding to the aligned extreme event; Based on the event covariance, the extreme covariance coefficient of the target stock is calculated.

9. The method for stock market monitoring and investment strategy optimization in a smart investment advisory system according to claim 1, wherein: The dynamically adjusting the position allocation weight of the target stock by combining the field volatility and the dynamic hidden danger infection entropy value includes: The position allocation weight of the target stock is dynamically adjusted using the following formula: ； Among them, G represents the position allocation weight of the target stock, M represents the volatility of the field, and N represents the dynamic hidden danger infection entropy value.

10. A stock market monitoring and investment strategy optimization system in a smart investment advisory system, characterized by: The system comprises: A module for generating an investment portfolio allocation tier is configured to obtain multi-dimensional market data of target stocks in the robo-advisory system in real time, analyze the financing scenario classification labels corresponding to the target stocks based on the multi-dimensional market data, collect the sector volatility and liquidity index of the target stocks, and generate an investment portfolio allocation tier for the target stocks based on the financing scenario classification labels, the sector volatility, and the liquidity index; A dynamic hidden danger transmission entropy value calculation module is used to calculate the volatility transmission coefficient of the target stock during the trading process, and to count the dynamic expenditure costs of the target stock during the holding period. The dynamic hidden danger transmission entropy value of the target stock is calculated by combining the volatility transmission coefficient and the dynamic expenditure costs; a strategy fitness evaluation module, configured to record the historical position change data and abnormal trading signals of the target stocks, calculate the position coupling deviation rate of the target stocks based on the historical position change data, calculate the extreme covariance coefficient of the target stocks based on the abnormal trading signals, and evaluate the strategy fitness between the target stocks by combining the position coupling deviation rate and the extreme covariance coefficient; A position optimization and control module is used to dynamically adjust the position allocation weight of the target stock in combination with the field volatility and the dynamic hidden danger infection entropy value, generate dynamic hedging instructions for the target stock according to the strategy fitness and the position allocation weight, and optimize and control the position of the target stock in real time based on the dynamic hedging instructions and the allocation tiers of the investment combination to obtain the combination adjustment strategy of the target stock.

Images