AI investment strategy auxiliary method and system

Abstract

The invention discloses an AI investment strategy auxiliary method and system. The method comprises the following steps: firstly, collecting and integrating market data to obtain a training set; then developing a data analysis model by using a neural network algorithm, and performing model training on the data analysis model through the training set and a manually labeled investment strategy auxiliary result; and finally, obtaining input data of a target user according to a pre-established user interface, and processing the input data by using the trained data analysis model to obtain an investment strategy auxiliary result of the target user. The investment strategy can be optimized in real time through the machine learning technology.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an AI investment strategy assistance method and system. Background Art

[0002] With the development of artificial intelligence technology, the financial investment field is increasingly dependent on intelligent decision support systems. Traditional investment strategies often rely on static data analysis and empirical judgment, which are difficult to adapt to the rapid changes in the market and the complexity of investor behavior. Prospect Theory and Cognitive Biases, as important theories of behavioral economics, provide new perspectives for understanding and predicting investor decision-making behavior. However, applying these theories to actual investment strategy optimization requires an intelligent system that can integrate multi-source data, dynamically learn, and adjust strategies.

[0003] There is an urgent need for a system that can integrate prospect theory and cognitive biases and optimize investment strategies in real time through machine learning technology. Summary of the invention

[0004] Based on this, the embodiments of the present application provide an AI investment strategy assistance method and system, which can optimize investment strategies in real time through machine learning technology.

[0005] In a first aspect, an AI investment strategy assistance method is provided, the method comprising:

[0006] Collecting and integrating market data to obtain a training set; wherein the market data includes historical volatility, GDP growth rate, unemployment rate and market sentiment index;

[0007] Developing a data analysis model using a neural network algorithm, and training the data analysis model using the training set and manually annotated investment strategy auxiliary results; wherein the data analysis model is used to analyze the decision-making patterns of investors under different market conditions, and the data analysis model at least includes probability weighting and value function;

[0008] The input data of the target user is obtained according to the pre-established user interface, and the input data is processed using the trained data analysis model to obtain the investment strategy auxiliary results of the target user.

[0009] Optionally, the method also includes: testing the data analysis model through the data set, receiving test feedback, and optimizing the system based on the feedback; adjusting functions according to user feedback, completing model optimization, and being responsible for deploying the model to the strategy output module.

[0010] Optionally, the method further includes: displaying the investment strategy assistance results of the target user through a visualization tool; wherein the investment strategy assistance results include various possibilities of investment decisions and their probabilities.

[0011] Optionally, the target user's investment strategy auxiliary results are displayed through a visualization tool, specifically including: using an exponential moving average EMA to provide the user with a data view, specifically including:

[0012] EMA t =α·x t +(1-α) EMA t-1

[0013] Among them, x t is the raw data point at time t; α is the smoothing factor, set between 0.1 and 0.3; EMA t is the smoothed output at time t.

[0014] Optionally, the collection and integration of market data also includes: performing sentiment analysis on market news through natural language processing technology to generate a market sentiment index.

[0015] Optionally, the data analysis model also includes a time variation factor so that the probability weighting function and value function in the model can be dynamically adjusted over time or other key events.

[0016] Optionally, a data analysis model is developed using a neural network algorithm, specifically including:

[0017] pass

[0018] NewOutput = Model(x, p, θ user )

[0019] Determine the investment strategy auxiliary result NewOutput, where Model is the data analysis model; x, p represent the characteristics and related probabilities of the investment plan respectively; θ user Represents a set of parameters adjusted by the user, which can introduce the effects of risk preference and time.

[0020] In a second aspect, an AI investment strategy assistance method and system are provided, the system comprising:

[0021] A data collection module, used to collect and integrate market data to obtain a training set; wherein the market data includes historical volatility, GDP growth rate, unemployment rate and market sentiment index;

[0022] A model training module, used to develop a data analysis model using a neural network algorithm, and to train the data analysis model using the training set and manually annotated investment strategy auxiliary results; wherein the data analysis model is used to analyze the decision-making patterns of investors under different market conditions, and the data analysis model at least includes probability weighting and value function;

[0023] The strategy output module is used to obtain the input data of the target user according to the pre-established user interface, and use the trained data analysis model to process the input data to obtain the investment strategy auxiliary results of the target user.

[0024] Optionally, the system also includes an optimization and testing module, which is used to test the data analysis model through the data set, receive test feedback, and optimize the system based on the feedback; adjust functions according to user feedback, complete model optimization, and be responsible for deploying the model to the strategy output module.

[0025] Optionally, the system further includes an output interface module for displaying the investment strategy assistance results of the target user through a visualization tool; wherein the investment strategy assistance results include various possibilities of investment decisions and their probabilities.

[0026] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least:

[0027] (1) By integrating prospect theory and cognitive biases and combining them with neural network algorithms, it is possible to more accurately simulate and predict investors’ decision-making patterns under different market conditions.

[0028] (2) Provide a clear and intuitive user interface that allows investors to easily operate and access real-time data and system-recommended investment decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0030] Figure 1 A flowchart of the steps of an AI investment strategy assistance method provided in an embodiment of the present application;

[0031] Figure 2 A block diagram of an AI investment strategy assistance system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0033] In the description of the present invention, the terms "comprises", "has" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may also include other steps or units that are not explicitly listed but are inherent to these processes, methods, products or apparatuses, or steps or units added based on further optimization schemes conceived by the present invention.

[0034] This system is a comprehensive and high-performance artificial intelligence support system designed for the investment decision-making process. It not only meets the current needs of investment management and decision-making support, improves the efficiency and accuracy of investment decisions, but also significantly optimizes investors' reactions and strategy choices when facing market fluctuations. In addition, by integrating the research results of prospect theory and cognitive bias, the system lays a solid foundation for the development of intelligent investment strategies, thereby promoting the advancement of financial technology and providing investors with excellent decision-making tools.

[0035] Please refer to Figure 1 , which shows a flow chart of an AI investment strategy assistance method provided by an embodiment of the present application, the method may include the following steps:

[0036] S1, collect and integrate market data to obtain the training set.

[0037] Among them, market data includes historical volatility, GDP growth rate, unemployment rate and market sentiment index.

[0038] S2, develops data analysis models using neural network algorithms, and trains the data analysis models using training sets and manually annotated investment strategy auxiliary results.

[0039] Among them, the data analysis model is used to analyze investors' decision-making patterns under different market conditions. The data analysis model at least includes probability weighting and value function.

[0040] S3, obtaining input data of the target user according to a pre-established user interface, and processing the input data using the trained data analysis model to obtain an investment strategy auxiliary result for the target user.

[0041] The method also includes testing the data analysis model through the data set, receiving feedback from the test, and optimizing the system based on the feedback; adjusting the function according to user feedback, completing the model optimization, and being responsible for deploying the model to the strategy output module.

[0042] The investment strategy assistance results of the target users are displayed through visualization tools; among which, the investment strategy assistance results include various possible investment decisions and their probabilities.

[0043] From the above, it can be seen that the various steps in the embodiments of this application are: first, develop a data analysis model that combines prospect theory and cognitive bias to accurately simulate and predict the behavior of investors in a changing market environment. Secondly, use machine learning technology to design an intelligent algorithm that dynamically learns and adjusts investment strategies to adapt to changes in real-time market data and historical investment behavior. Finally, develop an interactive software application that provides a clear and intuitive user interface so that investors can easily access and operate the system, thereby making more accurate and efficient investment decisions in the financial market. By introducing advanced technical solutions, it aims to improve the scientificity, convenience and effectiveness of investment decisions.

[0044] We seek innovations in the existing specific models of prospect theory and cognitive bias to better adapt to the application of AI investment decision-making optimization systems. We are now transforming and expanding the models through the following aspects.

[0045] (1) Multi-factor decision-making: Expand the traditional prospect theory model and incorporate more behavioral economics parameters and market variables into the model. For example, in addition to considering probability weighting and value functions, market sentiment, historical volatility, macroeconomic factors, etc. are introduced as decision variables. Among them, the market sentiment index is generated based on the sentiment analysis of market news, and the sentiment index is used as one of the inputs to reflect the impact of the overall market sentiment. In addition, macroeconomic factors, such as GDP growth rate, unemployment rate, inflation rate and other data, these macroeconomic indicators directly affect the investment environment and should be included in the decision-making model.

[0046] Assuming that x is a vector containing multiple market variables, such as the market sentiment index E, the macroeconomic factor M, and other relevant financial indicators, then this application can know that the combined effect of market sentiment and macroeconomic factors is:

[0047] V(x)=π(p)·v(x,E,M)

[0048] Among them, v(x, E, M) can be expressed as:

[0049] v(x,E,M)=(x α +δ E +δ M M)·λ

[0050] Among them, δ E , δ Mare the coefficients of the impact of market sentiment and macroeconomic factors on the value of investment decisions; λ is the loss aversion coefficient defined above.

[0051] (2) Dynamically adjusted risk preferences: People’s risk preferences are not static, but will change dynamically based on personal experience, market changes and other factors. This behavior is simulated by introducing a dynamic adjustment mechanism. In particular, a time change factor is introduced into the model so that the probability weighting function and the value function are dynamically adjusted over time or other key events. In addition, the key parameters in the model can be continuously updated through the learning and feedback mechanism of the AI ​​system, such as the probability weighting γ parameter and the integration of the α, β, and γ parameters of the value function. Assuming that the time variable t affects the parameters α, β, γ, and λ, they can be expressed as a function of time as follows:

[0052] α(t) represents the function of return sensitivity changing over time.

[0053] β(t) represents the function of loss sensitivity changing with time.

[0054] γ(t) represents the function of loss aversion coefficient changing with time.

[0055] λ(t) represents the function of the probability weighting exponent changing with time.

[0056] The above functions can be functions based on some external indicators (such as market volatility index), or they can be model parameters learned based on historical data. For example, the time changes of these parameters can be simulated by using a logistic function or an exponential decay function, that is, the following is obtained:

[0057]

[0058] Among them, α 0 , α 1 , k, t 0 As a parameter, it can be obtained by fitting historical data.

[0059] (3) Interactive simulation and prediction of market reactions: In investment decisions, it is necessary not only to consider the decisions of individual investors, but also to predict and simulate the reactions of other investors in the market and the impact of these reactions on the market. First, a multi-agent system is constructed, and each agent uses the optimization model of this application to make decisions. By simulating the interaction of the entire market, market dynamics and potential trends are predicted; secondly, how a single decision has an impact at the market level is analyzed, and this feedback is incorporated into the individual decision model to form a closed loop. Assume that the market consists of N agents, and the decision Dj of each agent j depends not only on his personal risk preference parameter θ j , and will also be affected by the average market decision, namely:

[0060] Dj (t) = f(θ j ,avg(D(t-1)),M(t))

[0061] Among them, f is the decision function, which is intended to include personal risk preference, the average decision of the market at the previous moment, and the current market information M(t); avg(D(t-1)) represents the average decision of the market at the previous moment; M(t) represents the market information at time t, such as market fluctuations, economic indicators, etc. In this way, the application can better understand and predict market behavior and provide investors with more accurate investment decision support.

[0062] (4) User-friendly interactive design is to enhance the practicality and accessibility of the model. Through user-friendly interface design, non-professional users can also understand and use the model to make investment decisions. First, develop intuitive visualization tools to display various possible outcomes of investment decisions and their probabilities to help users make more informed choices; second, allow users to set model parameters according to personal preferences, such as risk tolerance, investment goals, etc.

[0063] First, from the perspective of real-time data processing for visualization, real-time data processing and feedback are essential for interactive user interfaces. Considering the real-time flow of data and the need for user interaction, an exponential moving average (EMA) is used to provide users with a smoothed, easy-to-interpret data view, as shown below:

[0064] EMA t =α·x t +(1-α) EMA t-1

[0065] Among them, x t is the raw data point at time t; α is the smoothing factor, set between 0.1 and 0.3; EMA t It is a smoothed output at time t. Therefore, it can be applied to the real-time update of key indicators such as return on investment and market sentiment index to help users make faster and more accurate decisions.

[0066] Second, consider the ability provided to users in terms of interactive parameter adjustment capabilities to adjust model parameters in real time and observe how these changes affect the investment decision recommendations, as shown below:

[0067] NewOutput = Model(x, p, θ user )

[0068] Where Model is the Ai decision model; x, p represent the characteristics and related probabilities of the investment plan respectively; θ user It represents the set of parameters adjusted by the user, which can introduce the risk preference λ, the influence of time t, etc.user , instantly display new output results, improve user participation and system transparency.

[0069] Third, create dynamic risk assessment charts by visualizing the risk-return analysis of various investment decisions, namely:

[0070] Risk_Reward=π(p)·τ(x)-Cost(x)

[0071] Among them, Cost(x) is the cost of investment, a variable that is dynamically adjusted by user input. Dynamic charts not only enable investors to see the intuitive relationship between potential returns and risks, but also allow them to make adjustments and decisions based on their own risk tolerance.

[0072] The technical approach and route will be implemented in four stages: First, market and behavioral research will be conducted to establish a theoretical framework based on prospect theory and cognitive bias by analyzing investors' behavioral data. Secondly, based on this framework, machine learning models will be developed to use advanced algorithms such as neural networks to process and predict market changes. Next, these models and algorithms will be integrated to develop core software applications and conduct preliminary tests in a controlled environment. Finally, system optimization and user testing will be carried out to adjust functions based on feedback to ensure the user-friendliness of the software interface and the overall performance of the system. The entire technical development process is designed to ensure the efficiency and accuracy of the system while providing an easy-to-operate user interface to enhance the user experience.

[0073] like Figure 2 , the embodiment of the present application also provides an AI investment strategy assistance method and system. The system includes:

[0074] The data collection module is used to collect and integrate market data to obtain a training set; the market data includes historical volatility, GDP growth rate, unemployment rate, and market sentiment index;

[0075] The model training module is used to develop a data analysis model using a neural network algorithm, and to train the data analysis model using a training set and manually annotated investment strategy auxiliary results; the data analysis model is used to analyze the decision-making patterns of investors under different market conditions, and the data analysis model at least includes probability weighting and value functions;

[0076] The strategy output module is used to obtain the input data of the target user according to the pre-established user interface, and use the trained data analysis model to process the input data to obtain the investment strategy auxiliary results of the target user.

[0077] In an optional embodiment of the present application, the system also includes an optimization and testing module, which is used to test the data analysis model through a data set, receive test feedback, and optimize the system based on the feedback; adjust functions according to user feedback, complete model optimization, and is responsible for deploying the model to the strategy output module.

[0078] In an optional embodiment of the present application, the system also includes an output interface module for displaying the investment strategy assistance results of the target user through a visualization tool; wherein the investment strategy assistance results include various possible investment decisions and their probabilities.

[0079] The AI ​​investment strategy assistance method and system provided in the embodiments of the present application are used to implement the above-mentioned AI investment strategy assistance method. For the specific limitations of the AI ​​investment strategy assistance method and system, please refer to the limitations of the AI ​​investment strategy assistance method above, which will not be repeated here. The various parts of the above-mentioned AI investment strategy assistance method and system can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the device in the form of hardware, or can be stored in the memory in the device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0080] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0081] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. An AI investment strategy assistance method, characterized in that: The method comprises: Collecting and integrating market data to obtain a training set; wherein the market data includes historical volatility, GDP growth rate, unemployment rate and market sentiment index; Developing a data analysis model using a neural network algorithm, and training the data analysis model using the training set and manually annotated investment strategy auxiliary results; wherein the data analysis model is used to analyze the decision-making patterns of investors under different market conditions, and the data analysis model at least includes probability weighting and value function; The input data of the target user is obtained according to the pre-established user interface, and the input data is processed using the trained data analysis model to obtain the investment strategy auxiliary results of the target user.

2. The method according to claim 1, characterized in that The method also includes: testing the data analysis model through the data set, receiving feedback from the test, and optimizing the system based on the feedback; adjusting the function according to user feedback, completing the model optimization, and being responsible for deploying the model to the strategy output module.

3. The method according to claim 1, characterized in that The method also includes: displaying the investment strategy assistance results of the target user through a visualization tool; wherein the investment strategy assistance results include various possibilities of investment decisions and their probabilities.

4. The method according to claim 3, characterized in that: The results of the target user's investment strategy are shown through visualization tools. The exponential moving average (EMA) is used to provide users with data views, including: EMA t =α·x t +(1-a)·EMA t-1 Among them, x t is the raw data point at time t; α is the smoothing factor, set between 0.1 and 0.3; EMA t is the smoothed output at time t.

5. The method according to claim 1, characterized in that The collection and integration of market data also includes: using natural language processing technology to conduct sentiment analysis on market news and generate a market sentiment index.

6. The method according to claim 1, characterized in that The data analysis model also includes time-varying factors, which enable the probability weighting function and value function in the model to be dynamically adjusted over time or other key events.

7. The method according to claim 1, characterized in that Develop data analysis models using neural network algorithms, including: pass NewOutput=Model(x,p,θ user ) Determine the investment strategy auxiliary result NewOutput, where Model is the data analysis model; x, p represent the characteristics and related probabilities of the investment plan respectively; θ user Represents a set of parameters adjusted by the user, which can introduce the effects of risk preference and time.

8. An AI investment strategy assistance system, characterized in that: The system comprises: A data collection module, used to collect and integrate market data to obtain a training set; wherein the market data includes historical volatility, GDP growth rate, unemployment rate and market sentiment index; A model training module, used to develop a data analysis model using a neural network algorithm, and to train the data analysis model using the training set and manually annotated investment strategy auxiliary results; wherein the data analysis model is used to analyze the decision-making patterns of investors under different market conditions, and the data analysis model at least includes probability weighting and value function; The strategy output module is used to obtain the input data of the target user according to the pre-established user interface, and use the trained data analysis model to process the input data to obtain the investment strategy auxiliary results of the target user.

9. The system according to claim 8, characterized in that The system also includes an optimization and testing module, which is used to test the data analysis model through data sets, receive test feedback, and optimize the system based on the feedback; adjust functions according to user feedback, complete model optimization, and is responsible for deploying the model to the strategy output module.

10. The system according to claim 8, characterized in that The system also includes an output interface module for displaying the investment strategy assistance results of the target user through a visualization tool; wherein the investment strategy assistance results include various possibilities of investment decisions and their probabilities.

Images