Machine transaction management method, system and equipment and storage medium

Abstract

The invention discloses a machine transaction management method, system and device and a storage medium, and relates to the technical field of transaction management, and the method comprises the steps: collecting the historical transaction behaviors of a user, analyzing the historical transaction behaviors of the user and the current market dynamic state through a deep learning model, automatically recommending a transaction strategy, and generating a virtual transaction machine instance; after self-defined parameters of the virtual transaction machine instance are set, transaction operation is automatically executed according to real-time data of the market; in the execution process, an optimal transaction execution path is found through a quantum computing technology, transaction risks are monitored in combination with a risk control mechanism, and a transaction strategy is adjusted; recording earnings, transaction details and shipping space changes of the virtual transaction machine instance, generating a performance report, and proposing a transaction strategy optimization suggestion and a fund management suggestion; by introducing deep learning and quantum computing technologies, the automation and intelligence level of quantitative transaction management is significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction management, and in particular to a management method, system, device and storage medium for machine transactions. Background Art

[0002] In recent years, quantitative trading technology has been widely used in financial markets and has made significant progress in automated trading, market analysis, and risk management. Many technical means, such as machine learning, statistical analysis, and algorithmic trading, have been used to improve the accuracy and efficiency of trading strategies. However, despite the progress of these technologies, existing quantitative trading still has shortcomings in strategy recommendation, automated execution, and risk control.

[0003] First, the generation of trading strategies often lacks personalization and cannot be adjusted in real time based on individual users' trading behaviors and market dynamics. Second, the optimization of trading execution paths is relatively limited, and it is impossible to fully utilize the latest computing technologies such as quantum computing to improve trading efficiency. In addition, the risk control mechanism is relatively simple and usually relies on preset static risk control parameters, which makes it difficult to cope with drastic market fluctuations. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a machine trading management method, system, device and storage medium to solve the problems of insufficient personalized strategy recommendation, insufficient optimization of transaction execution path and limited dynamic risk control capability.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for managing machine trading, which includes collecting historical trading behaviors of users, analyzing historical trading behaviors of users and current market dynamics using a deep learning model, automatically recommending trading strategies, and generating virtual trading machine instances;

[0008] After setting custom parameters of the virtual trading machine instance, automatically perform trading operations based on real-time market data;

[0009] During the execution process, the optimal transaction execution path is found through quantum computing technology, and the transaction risk is monitored and the transaction strategy is adjusted in combination with the risk control mechanism;

[0010] Record the profits, transaction details and position changes of virtual trading machine instances, generate performance reports, and provide trading strategy optimization suggestions and fund management suggestions.

[0011] As a preferred solution of the method for managing machine transactions of the present invention, wherein: the historical transaction behavior includes transaction frequency, transaction size and risk preference;

[0012] The current market dynamics include price action, volume, market trends and volatility.

[0013] As a preferred solution of the management method of machine trading described in the present invention, the deep learning model is used to analyze the user's historical trading behavior and current market dynamics, automatically recommend trading strategies, and generate a virtual trading machine instance, including the following steps:

[0014] Load historical trading strategies and other successful users' trading strategies from the user database;

[0015] Use long short-term memory networks to combine users' historical trading behaviors and current market dynamics to generate user profiles and identify users' trading strategies;

[0016] By comparing the user's trading strategy with the trading strategy choices of other successful users, the collaborative filtering algorithm is used to recommend the most matching trading strategy;

[0017] Create a virtual trading machine instance based on the recommended best-matching trading strategy.

[0018] As a preferred solution of the management method of machine trading of the present invention, after setting the custom parameters of the virtual trading machine instance, the trading operation is automatically executed according to the real-time data of the market, including the following steps:

[0019] Recommend the most suitable contract for the current market based on the user's historical contract selection;

[0020] Recommend directions to users based on current market trends;

[0021] By using the Kelly formula, the initial amount of funds is recommended based on the user's historical fund usage, risk appetite, and current market volatility;

[0022] Set the opening price and the initial opening lot size based on the current market price and volatility;

[0023] According to the set custom parameters and combined with the real-time market data, a virtual trading machine instance is started, and based on the real-time market data dynamics, the market price trend is captured and the trading signals are calculated. When the trading signal is triggered, the trading operation is automatically executed according to the trading strategy.

[0024] As a preferred solution of the management method of machine transactions described in the present invention, in the execution process, the optimal transaction execution path is found through quantum computing technology, and the transaction risk is monitored and the transaction strategy is adjusted in combination with the risk control mechanism, including the following steps:

[0025] Using quantum computing technology, the user's trading strategy and real-time market data are used as input, and multiple trading execution paths are generated through quantum Monte Carlo simulation to quantify the benefits and risks of each path;

[0026] Based on the evaluation results, the transaction path with the best balance between benefits and risks is selected as the optimal transaction execution path;

[0027] Load the risk control parameters set by the user and set them as risk control mechanisms;

[0028] Monitor market dynamics in real time through risk control mechanisms, and assess the risk status at each moment based on the current transaction execution path;

[0029] Combine the assessed risk status with the market's price trend to dynamically adjust trading strategies.

[0030] As a preferred solution of the management method of machine trading described in the present invention, the following steps are included: recording the income, transaction details and position changes of the virtual trading machine instance, generating a performance report, and proposing trading strategy optimization suggestions and fund management suggestions.

[0031] Record the actual transaction price, transaction time, holding days and transaction fees during the transaction process to facilitate the review and optimization of trading strategies;

[0032] Record the details of each transaction;

[0033] For each transaction, record the position changes of the virtual trading machine instance, including initial position, position reduction or increase, and position closing;

[0034] Analyze the profits, transaction details and position changes during the transaction process to obtain a performance report;

[0035] Retrain users' trading strategies through deep learning models, combine historical trading behaviors and current market dynamics, identify users' successful and failed strategies under different market conditions, and make suggestions for strategy adjustments based on the model's prediction results;

[0036] Based on the user's historical win rate, profit and loss ratio and risk preference, the Kelly formula is used to recalculate the optimal fund allocation ratio, generate a fund management report, display the user's historical fund utilization efficiency, and make recommendations on fund allocation.

[0037] In a second aspect, the present invention provides a quantitative trading management system, including a strategy recommendation and virtual trading machine generation module, which is used to collect historical trading behaviors of users, analyze historical trading behaviors of users and current market dynamics using a deep learning model, automatically recommend trading strategies, and generate virtual trading machine instances;

[0038] The trading execution module is used to automatically execute trading operations based on real-time market data after setting custom parameters of the virtual trading machine instance;

[0039] The quantum computing and risk control module is used to find the optimal transaction execution path through quantum computing technology during the execution process, and monitor transaction risks and adjust transaction strategies in combination with risk control mechanisms;

[0040] The performance evaluation and optimization suggestion module is used to record the income, transaction details and position changes of the virtual trading machine instance, generate performance reports, and put forward trading strategy optimization suggestions and fund management suggestions.

[0041] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for managing machine transactions as described in the first aspect of the present invention is implemented.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the method for managing machine transactions as described in the first aspect of the present invention.

[0043] The beneficial effects of the present invention are as follows: by introducing deep learning and quantum computing technologies, the automation and intelligence level of quantitative trading management is significantly improved. First, the deep learning model can comprehensively analyze the user's historical trading behavior and current market dynamics, and generate personalized trading strategies, thereby improving the accuracy and pertinence of strategy recommendations. During the transaction execution process, quantum computing technology is used to optimize the transaction execution path, quantify and balance the risks and benefits of each path, and ensure that the optimal transaction path is selected in a complex market environment. In addition, by combining the risk control mechanism, the present invention can monitor market dynamics in real time, automatically adjust trading strategies and position ratios, and ensure that trading risks are within a controllable range, especially in the case of large market volatility, effectively reducing potential losses. Through a series of improvements, the present invention not only improves the user's rate of return, but also significantly improves the robustness of transactions and risk management capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0045] Figure 1This is a flowchart of the method for managing machine transactions in Example 1.

[0046] Figure 2 This is a flow chart for evaluating risk status in Example 1. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0050] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a method for managing machine transactions, comprising the following steps:

[0051] S1. Collect the user's historical trading behavior, use the deep learning model to analyze the user's historical trading behavior and current market dynamics, automatically recommend trading strategies, and generate a virtual trading machine instance, including the following steps:

[0052] Historical trading behavior including trading frequency, trading size, and risk appetite;

[0053] Current market dynamics include price action, volume, market trends and volatility.

[0054] Load historical trading strategies and other successful users' trading strategies from the user database;

[0055] Use long short-term memory networks to combine users' historical trading behaviors and current market dynamics to generate user profiles and identify users' trading strategies;

[0056] Specifically, the LSTM network (Long Short-Term Memory Network) model is used to process time series data to capture the long-term trend and short-term fluctuation of user trading behavior. The trading behavior sequence (such as trading strategy, opening / closing time, profit and loss situation) is input, and the user portrait is output to represent the user's trading characteristics (such as risk preference, trading style, etc.). The current market dynamics (such as current price, trading volume, volatility, market sentiment, etc.) are introduced as one of the inputs of the LSTM model, so that the user portrait is not only based on historical data, but also reflects the current market conditions. The current market dynamics data is input into the LSTM model as a time series and combined with the user's historical trading behavior to generate a real-time updated user portrait. The combination of current market data and historical trading behavior can help the model capture the user's trading strategy under different market conditions;

[0057] By comparing the user's trading strategy with the trading strategy choices of other successful users, the collaborative filtering algorithm is used to recommend the most matching trading strategy;

[0058] It should be noted that collaborative filtering can recommend unused trading strategies by calculating similarities based on a user-strategy matrix (e.g., user A uses strategy X, user B uses strategy Y). Based on the similarities calculated by the collaborative filtering algorithm, the most matching trading strategies are recommended, with priority given to the trading strategies adopted by successful users who are most similar to the current user profile.

[0059] Based on the recommended best-matching trading strategy, a virtual trading machine instance is created, including default trading parameters and interface methods.

[0060] It should be noted that by collecting the user's historical trading behavior, using the deep learning model to analyze the user's historical trading behavior and current market dynamics, automatically recommending trading strategies, and generating virtual trading machine instances, it is possible to combine the user's historical trading data (such as trading frequency, trading scale, and risk preference) and current market dynamics (such as price trends, trading volume, market trends, and volatility) to generate a real-time feedback user portrait. Then, based on these portraits, the deep learning model automatically recommends the most matching trading strategy and generates a virtual trading machine instance. The significant role of this automated process is to improve the personalization and pertinence of trading strategies, avoid the limitations of traditional quantitative trading that lack dynamic adjustments, and thus achieve the accuracy and market adaptability of strategy recommendations. Ultimately, this step greatly improves the success probability and efficiency of user transactions.

[0061] S2. After setting the custom parameters of the virtual trading machine instance, the trading operation is automatically executed according to the real-time data of the market, including the following steps:

[0062] Set the machine name, for example, AAPL_Trend Following_Long. Users can choose to modify the name to personalize the instance and ensure that users can quickly identify it among multiple instances.

[0063] Recommend the most suitable contract for the current market based on the user's historical contract selection;

[0064] Recommend direction (long or short) to users based on current market dynamics (such as market trends, volatility, trading volume, etc.). If the market is in a strong upward trend, a long direction may be recommended;

[0065] Calculate the trend strength indicator, which can be used to determine the market trend, expressed as,

[0066]

[0067] Among them, T represents the trend strength indicator, which is used to identify the upward or downward trend of the market. Users can decide the direction of buying and selling based on it. A represents the operation of smoothing price data, e represents the price of the asset, and r1 and r2 represent different period parameters;

[0068] By using the Kelly formula, the initial amount of funds is recommended based on the user's historical fund usage, risk preference and current market volatility;

[0069] Specifically, the Kelly formula can be used to recommend the optimal capital allocation ratio based on the user's historical win rate and profit-loss ratio, expressed as:

[0070]

[0071] Among them, f * It represents the recommended capital allocation ratio, which is used to calculate the amount of capital allocated to the strategy. b represents the profit-loss ratio, p represents the winning rate, and q represents the probability of loss.

[0072] Set the opening price and the initial opening lot size based on the current market price and volatility;

[0073] It should be noted that the calculation formula for the number of opening positions is expressed as:

[0074]

[0075] Among them, N represents the recommended opening lot size, which can help users reasonably allocate the opening lot size under different market conditions to ensure that the risk is within a controllable range. E represents the account funds, F represents the risk factor, V represents the current contract price, and σ represents the market volatility.

[0076] According to the set custom parameters, combined with the real-time market data, including price, trading volume, market depth, volatility and other information, the virtual trading machine instance is started, and based on the real-time market data dynamics, the market price trend is captured, and trading signals (such as buy and sell, etc.) are calculated. When the trading signal is triggered, the trading operation is automatically executed according to the trading strategy. The buy signal is triggered when the price breaks through a key resistance level and meets the strategy conditions. The sell signal is triggered when the price falls below a support level or reaches the profit target. The closing signal is to close the position according to the set stop loss, take profit conditions, or strategy trigger conditions;

[0077] It should be noted that to capture the price trend of the market, we can calculate two moving averages of different periods, take the moving average as the center line, and calculate the upper and lower tracks based on the price volatility. When the price touches or falls below the lower track of the Bollinger Band, a buy signal is generated, assuming that the price will return to the middle line. When the price touches or breaks through the upper track of the Bollinger Band, a sell signal is generated, assuming that the price will return to the middle line.

[0078] It should be noted that by allowing users to set custom parameters of virtual trading machine instances (such as machine name, trading contract, opening price, etc.), the trading machine is highly configurable. Combined with real-time market data (such as trading volume, volatility and market trends), the virtual machine can automatically perform trading operations such as buying, selling and closing positions. The key role of this operation is to greatly reduce manual intervention and ensure that trading operations can respond quickly to market changes, especially in high-frequency or high-volatility markets. This can effectively reduce the risks caused by delays, while also improving the flexibility and real-time nature of users' trading operations. Ultimately, trading efficiency and execution accuracy have been significantly improved.

[0079] S3. During the execution process, the optimal transaction execution path is found through quantum computing technology, and the transaction risk is monitored and the transaction strategy is adjusted in combination with the risk control mechanism, including the following steps:

[0080] Using quantum computing technology, the user's trading strategy and real-time market data (such as price, volume, volatility, etc.) are used as input to generate multiple trading execution paths through quantum Monte Carlo simulation, and quantify the benefits and risks of each path;

[0081] It should be noted that trading risk is to quantify the risk of each path by combining market volatility and the user's risk preference, while quantified returns are to simulate the returns under different market trends;

[0082] Based on the evaluation results, the transaction path with the best balance between benefits and risks is selected as the optimal transaction execution path;

[0083] Load the risk control parameters set by the user, including stop loss, take profit, maximum drawdown, etc., and set them as risk control mechanisms;

[0084] Monitor market dynamics in real time through risk control mechanisms, and assess the risk status at each moment based on the current transaction execution path;

[0085] Specifically, the risk value is calculated using the current account's drawdown value and the maximum drawdown value set by the user, expressed as:

[0086]

[0087] Among them, R represents the risk value, C represents the drawdown value of the current account, and M represents the maximum drawdown value set by the user;

[0088] Calculate the average risk value over a period of time as the risk threshold. When the risk value exceeds the risk threshold, issue an alarm and prepare to adjust the trading strategy. If it does not exceed the risk threshold, continue to monitor market dynamics.

[0089] Combine the assessed risk status with the market price trend to dynamically adjust the trading strategy;

[0090] It should be noted that if the market volatility increases, the position ratio will be automatically reduced according to the quantum computing results to reduce risks. If the market trend changes, the risk control mechanism will dynamically adjust the stop loss or take profit points. If the market trend weakens, the position may be automatically closed or reduced; if the trend strengthens, the position may be increased.

[0091] Furthermore, when the virtual trading machine instance is paused or stopped, the user can pause the machine's operation at any time. In the paused state, the machine maintains the current position and order information but does not execute new trading operations. Quantum computing stops optimizing calculations and saves market data and transaction records for future recovery. When the user decides to stop the machine, its trading permissions are closed and historical transaction records are saved for subsequent queries.

[0092] It should be noted that during the execution process, the optimal transaction execution path is found through quantum computing technology, and the risk control mechanism is combined to monitor transaction risks, adjust trading strategies, and use quantum Monte Carlo simulation technology to analyze users' trading strategies and market data, and quickly generate and evaluate multiple possible transaction execution paths. This process not only optimizes the balance between returns and risks, but also greatly improves the calculation efficiency of transaction paths. Combined with the risk control mechanism, real-time monitoring of market dynamics (such as volatility, maximum drawdown, etc.), dynamically adjust trading strategies according to risk status, such as reducing positions or adjusting stop loss points. The role of this step is reflected in the ability to dynamically respond to risks under complex market conditions, and to protect the user's financial security to the greatest extent, ensuring the robustness of transactions and the sustainability of returns.

[0093] S4. Record the income, transaction details and position changes of the virtual trading machine instance, generate a performance report, and put forward suggestions for trading strategy optimization and fund management, including the following steps:

[0094] Record the actual transaction price, transaction time, holding days and transaction fees during the transaction process to facilitate the review and optimization of trading strategies;

[0095] Record detailed information of each transaction, including transaction direction (buy / sell), transaction type (such as AAPL, SPY, etc.), opening price, closing price, opening lot number, closing lot number, transaction profit and loss (i.e. PnL), transaction time (opening time and closing time), transaction volume and other market dynamic information (such as volatility, transaction volume);

[0096] For each transaction, record the position changes of the virtual trading machine instance, including initial position, position reduction or increase, and position closing;

[0097] Analyze the profits, transaction details and position changes during the transaction process to obtain a performance report;

[0098] Specifically, it calculates the user's overall rate of return, net income, annualized rate of return, maximum drawdown, Sharpe ratio, Karma ratio and other important performance indicators during the statistical period, summarizes these indicators into a report, and displays the distribution, trend and fluctuation of income in the form of charts and tables to help users intuitively understand the performance of trading strategies; displays the profit and loss of each transaction separately, lists the detailed information of each transaction in chronological order, and can generate a profit and loss distribution chart to display the profit and loss distribution of the transaction, helping users understand which trading strategies or market scenarios have the best performance and in which situations the losses are more serious; analyzes the relationship between position changes and market fluctuations, and generates a correlation analysis chart between positions and market volatility to help users judge whether position management is reasonable; reviews the risk control effect throughout the entire trading cycle, analyzes the triggering of each stop loss and stop profit, calculates the average stop loss points, average stop profit points and trigger frequency, generates relevant statistical charts, and generates a maximum drawdown analysis report to help users understand the risk exposure under extreme market conditions and how to reduce drawdowns by adjusting strategies, integrates all reports and charts, and generates performance reports;

[0099] Retrain the user's trading strategy through deep learning models (such as LSTM networks), combine historical trading behavior and current market dynamics, identify the user's successful and failed strategies under different market conditions, and make strategy adjustment suggestions based on the model's prediction results;

[0100] If the model detects that the user's trading frequency is too high and the profit effect is not good, it is recommended to reduce the number of transactions and focus on higher-quality trading opportunities. By reviewing historical trading performance, it recommends more appropriate stop loss and take profit points to reduce premature stop loss or miss the best take profit point. If the position allocation does not match the market volatility, it is recommended to reduce the position in the high-volatility market and increase the position in the low-volatility market;

[0101] Based on the user's historical win rate, profit-loss ratio and risk preference, the Kelly formula is used to recalculate the optimal fund allocation ratio to help users reasonably allocate funds to different trading strategies or markets, generate fund management reports, display the user's historical fund utilization efficiency, and make suggestions on fund allocation to maximize returns while controlling risks;

[0102] If it is detected that the user's funds are too concentrated in a certain market or trading strategy, we will make suggestions for diversification of investments. For example, we will suggest diversifying the funds into different products, different markets (such as stocks, futures, foreign exchange, etc.) or different time periods (such as short-term, medium-term, and long-term strategies). Based on the current market dynamics, we will recommend several different trading products and trading strategy combinations to help users balance risks and returns.

[0103] It should be noted that by recording the income, transaction details and position changes of the virtual trading machine instance, generating a performance report, and proposing trading strategy optimization suggestions and fund management suggestions, the detailed information of the trading operation can be recorded throughout the process, including the direction, type, opening and closing prices, number of lots and profit and loss of each transaction. After analysis, these data generate a comprehensive performance report, showing the user's key indicators such as rate of return, maximum drawdown, Sharpe ratio, and further proposing strategy optimization suggestions and fund management suggestions through deep learning models. The core role of this step is to provide users with clear trading performance feedback, and help users improve future trading strategies and fund allocation through data-driven decision optimization guidance. Ultimately, this step not only improves the user's trading strategy optimization capabilities, but also provides a scientific basis for longer-term fund management.

[0104] This embodiment also provides a quantitative trading management system, including: a strategy recommendation and virtual trading machine generation module, which is used to collect historical trading behaviors of users, analyze historical trading behaviors of users and current market dynamics using a deep learning model, automatically recommend trading strategies, and generate virtual trading machine instances;

[0105] The trading execution module is used to automatically execute trading operations based on real-time market data after setting custom parameters of the virtual trading machine instance;

[0106] The quantum computing and risk control module is used to find the optimal transaction execution path through quantum computing technology during the execution process, and monitor transaction risks and adjust transaction strategies in combination with risk control mechanisms;

[0107] The performance evaluation and optimization suggestion module is used to record the income, transaction details and position changes of the virtual trading machine instance, generate performance reports, and put forward trading strategy optimization suggestions and fund management suggestions.

[0108] This embodiment also provides a computer device suitable for the case of a method for managing machine transactions, comprising: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method for managing machine transactions proposed in the above embodiment.

[0109] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0110] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the management method for implementing machine transactions as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0111] In summary, the present invention significantly improves the automation and intelligence level of quantitative trading management by introducing deep learning and quantum computing technology. First, the deep learning model can comprehensively analyze the user's historical trading behavior and current market dynamics, generate personalized trading strategies, and thus improve the accuracy and pertinence of strategy recommendations. During the transaction execution process, quantum computing technology is used to optimize the transaction execution path, quantify and balance the risks and benefits of each path, and ensure that the optimal transaction path is selected in a complex market environment. In addition, by combining the risk control mechanism, the present invention can monitor market dynamics in real time, automatically adjust trading strategies and position ratios, and ensure that trading risks are within a controllable range, especially in the case of large market volatility, effectively reducing potential losses. Through a series of improvements, the present invention not only improves the user's rate of return, but also significantly improves the robustness of transactions and risk management capabilities.

[0112] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a method for managing machine transactions are provided.

[0113] In this experiment, the experimental group adopted a management method for machine trading based on deep learning models and quantum computing technology, which mainly includes analyzing the user's historical trading behavior and current market dynamics through LSTM (Long Short-Term Memory Network), automatically recommending trading strategies, and generating virtual trading machine instances; then, after setting custom parameters, automatically executing trading operations based on real-time market data, and finally finding the optimal trading execution path through quantum computing technology, and adjusting trading strategies in combination with risk control mechanisms. The trading strategy generation process of the experimental group utilized user portraits, combined with the user's historical data and real-time market dynamics, to ensure that the recommended strategy is highly matched with the current market environment.

[0114] The control group used a traditional quantitative trading method to generate trading strategies based on historical market data and fixed rules, without using deep learning or quantum computing technology. The virtual trading machine instances in the control group rely on fixed parameter settings and do not have the ability to adjust to real-time market dynamics. All transactions are conducted in the same market environment, and the market data comes from the futures market in the third quarter of 2023.

[0115] The key parameters of the experiment include trading frequency (number of transactions per month), trading size (amount of funds per transaction), trading win rate (success rate of transactions), annualized rate of return, maximum drawdown and Sharpe ratio. The experimental data was obtained by simulating the trading process in the real market environment, and the experimental time was three months.

[0116] The details are shown in Table 1 below:

[0117] Table 1 Experimental record table

[0118]

[0119] By comparing the data of the experimental group and the control group in Table 1, we can clearly see the significant advantages of deep learning and quantum computing technology in trading strategy optimization and execution. The trading frequency of the experimental group is 25 times / month, which is much higher than the 12 times / month of the control group, indicating that the experimental group can capture market opportunities more frequently. This is due to the fact that the virtual trading machine instance of the experimental group can analyze market dynamics in real time and automatically adjust trading strategies, thereby improving the flexibility and real-time nature of trading.

[0120] In terms of transaction size, the average amount of funds per transaction in the experimental group was US$50,000, 67% more than the US$30,000 in the control group. This reflects that the fund allocation calculated by the Kelly formula in the experimental group is more scientific and reasonable, and can maximize the efficiency of fund use under the premise of controllable risks. The transaction winning rate of the experimental group is 72.5%, much higher than the 55.3% of the control group, which shows that the user profiling and dynamic strategy generation based on deep learning can more effectively consider market fluctuations and user risk preferences, and recommend more accurate trading strategies.

[0121] In terms of annualized rate of return, the experimental group reached 24.8%, almost twice the 12.4% of the control group. Combined with the data analysis of the Sharpe ratio (1.65 vs 0.92) and the maximum drawdown (8.1% vs 15.6%), the experimental group not only has higher returns, but also more effective risk control. The maximum drawdown of the experimental group is only 8.1%, significantly lower than the 15.6% of the control group. This shows that the experimental group can effectively reduce losses when dealing with extreme market fluctuations, thanks to the combination of quantum computing technology and risk control mechanisms. In addition, the Sharpe ratio of the experimental group is 1.65, indicating that it obtains higher returns per unit of risk, reflecting its superiority in risk management.

[0122] In general, it can be seen from the experimental data that the management method of machine trading of the present invention is significantly superior to traditional methods in terms of trading frequency, capital allocation, risk control and returns, showing innovation and significant market adaptability. This quantitative trading based on deep learning and quantum computing technology can not only better capture market opportunities, but also achieve higher returns with higher winning rates and lower risks, fully demonstrating its application value in a complex market environment.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. 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, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for managing machine transactions, characterized in that: include, Collect users' historical trading behaviors, use deep learning models to analyze users' historical trading behaviors and current market dynamics, automatically recommend trading strategies, and generate virtual trading machine instances; After setting custom parameters of the virtual trading machine instance, automatically perform trading operations based on real-time market data; During the execution process, the optimal transaction execution path is found through quantum computing technology, and the transaction risk is monitored and the transaction strategy is adjusted in combination with the risk control mechanism; Record the profits, transaction details and position changes of virtual trading machine instances, generate performance reports, and provide trading strategy optimization suggestions and fund management suggestions.

2. The method for managing machine transactions according to claim 1, characterized in that: The historical trading behavior includes trading frequency, trading size and risk appetite; The current market dynamics include price action, volume, market trends and volatility.

3. The method for managing machine transactions according to claim 2, characterized in that: Using deep learning models to analyze users’ historical trading behaviors and current market dynamics, automatically recommend trading strategies, and generate virtual trading machine instances, includes the following steps: Load historical trading strategies and other successful users' trading strategies from the user database; Use long short-term memory networks to combine users' historical trading behaviors and current market dynamics to generate user profiles and identify users' trading strategies; By comparing the user's trading strategy with the trading strategy choices of other successful users, the collaborative filtering algorithm is used to recommend the most matching trading strategy; Create a virtual trading machine instance based on the recommended best-matching trading strategy.

4. The method for managing machine transactions according to claim 3, characterized in that: After setting the custom parameters of the virtual trading machine instance, the trading operations are automatically performed according to the real-time data of the market, including the following steps: Recommend the most suitable contract for the current market based on the user's historical contract selection; Recommend directions to users based on current market trends; By using the Kelly formula, the initial amount of funds is recommended based on the user's historical fund usage, risk appetite, and current market volatility; Set the opening price and the initial opening lot size based on the current market price and volatility; According to the set custom parameters and combined with the real-time market data, a virtual trading machine instance is started, and based on the real-time market data dynamics, the market price trend is captured and the trading signals are calculated. When the trading signal is triggered, the trading operation is automatically executed according to the trading strategy.

5. The method for managing machine transactions according to claim 4, characterized in that: During the execution process, the optimal transaction execution path is found through quantum computing technology, and the transaction risk is monitored and the transaction strategy is adjusted in combination with the risk control mechanism, including the following steps: Using quantum computing technology, the user's trading strategy and real-time market data are used as input, and multiple trading execution paths are generated through quantum Monte Carlo simulation to quantify the benefits and risks of each path; Based on the evaluation results, the transaction path with the best balance between benefits and risks is selected as the optimal transaction execution path; Load the risk control parameters set by the user and set them as risk control mechanisms; Monitor market dynamics in real time through risk control mechanisms, and assess the risk status at each moment based on the current transaction execution path; Combine the assessed risk status with the market's price trend to dynamically adjust trading strategies.

6. The method for managing machine transactions according to claim 5, characterized in that: Record the earnings, transaction details and position changes of the virtual trading machine instance, generate performance reports, and make suggestions for trading strategy optimization and fund management, including the following steps: Record the actual transaction price, transaction time, holding days and transaction fees during the transaction; Record the details of each transaction; For each trade, the position changes of the virtual trading machine instance are recorded; Analyze the profits, transaction details and position changes during the transaction to obtain a performance report; Retrain users' trading strategies through deep learning models, combine historical trading behaviors and current market dynamics, identify users' successful and failed strategies under different market conditions, and make suggestions for strategy adjustments based on the model's prediction results; Based on the user's historical win rate, profit and loss ratio and risk preference, the Kelly formula is used to recalculate the optimal fund allocation ratio, generate a fund management report, and make recommendations on fund allocation.

7. A quantitative trading management system, based on the management method of machine trading according to any one of claims 1 to 6, characterized in that: include, Strategy recommendation and virtual trading machine generation module, which is used to collect users' historical trading behaviors, analyze users' historical trading behaviors and current market dynamics using deep learning models, automatically recommend trading strategies, and generate virtual trading machine instances; The trading execution module is used to automatically execute trading operations based on real-time market data after setting custom parameters of the virtual trading machine instance; The quantum computing and risk control module is used to find the optimal transaction execution path through quantum computing technology during the execution process, and monitor transaction risks and adjust transaction strategies in combination with risk control mechanisms; The performance evaluation and optimization suggestion module is used to record the income, transaction details and position changes of the virtual trading machine instance, generate performance reports, and put forward trading strategy optimization suggestions and fund management suggestions.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for managing machine transactions according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for managing machine transactions according to any one of claims 1 to 6 are implemented.

Images