Multi-index analysis-based constant future quantitative transaction system and method

Abstract

The invention relates to the technical field of finance, and discloses a multi-index analysis-based constant-index future quantitative transaction system, which comprises a data acquisition module for acquiring real-time market data from a constant-index future transaction platform, including one-minute K-line data, opening price, maximum price, minimum price, closing price and trading volume; the data processing module is used for caching the obtained market data into a memory and updating the market data in real time, and caching latest 6000 pieces of K line data by adopting a double-end queue structure; and the technical index calculation module is used for calculating a plurality of technical indexes such as Kalman filtering, Bayesian inference, E40 resonance, E41 resonance and a Boforest belt. By combining multiple technical indexes, the market trend, volatility and inversion signals are accurately captured, so that the accuracy of transaction decision making is improved. And the system adopts automatic decision execution and risk control, so that the bin position and the stop loss and interference point can be adjusted in real time, and the risk caused by manual operation is reduced.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and in particular to a Hang Seng Index futures quantitative trading system and method based on multi-index analysis. Background Art

[0002] Existing automated trading systems mostly rely on traditional technical indicators for market analysis, such as Bollinger Bands, RSI, MACD, etc. Although these technical indicators are widely used in market analysis, they have certain limitations when dealing with complex and rapidly changing markets. A single technical indicator can only provide partial market information and cannot fully capture the market's volatility, trends, and reversal signals. When the market changes rapidly, these indicators often lag behind, resulting in the inability to make correct trading decisions in a timely manner. Therefore, traditional technical solutions that rely on a single indicator are difficult to adapt to changing market demands.

[0003] Many existing automated trading systems use fixed strategies and parameters. Although this approach is effective under certain market conditions, it lacks flexibility and is difficult to cope with changes in the market environment. For example, market trends may reverse quickly, and the performance of volatile markets and trending markets varies greatly. Traditional fixed strategies often fail to work as expected under different market conditions. Fixed strategies may lead to unnecessary stop losses in volatile markets or missed profit opportunities in trending markets. Therefore, fixed strategies lack adaptability, cannot be optimized and adjusted in real time, and cannot fully adapt to dynamic changes in the market.

[0004] In addition, the existing trading system has a certain lag in data processing, especially in high-frequency trading scenarios, where the real-time and update frequency of data are crucial. The caching mechanism and data update method used in the existing system have failed to effectively solve the problem of real-time data transmission and processing. Data delay will directly affect the generation and execution of trading signals and miss the best market opportunities. Especially when the market fluctuates rapidly, data delay will lead to wrong trading decisions or failure to execute trading instructions in time, further affecting the stability and profitability of trading.

[0005] Therefore, existing technologies have obvious shortcomings in accurately capturing market information, dynamically adjusting trading strategies, and efficiently processing real-time data. A more comprehensive and flexible technical solution is needed to solve these problems. Summary of the invention

[0006] In view of the deficiencies of the prior art, the present invention provides a Hang Seng Index futures quantitative trading system and method based on multi-indicator analysis, which solves the problems of single indicator analysis, lack of flexible strategy adjustment mechanism and data delay in the existing automated trading system.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a Hang Seng Index futures quantitative trading system based on multi-index analysis, including a data acquisition module for acquiring real-time market data from the Hang Seng Index futures trading platform, including 1-minute K-line data, opening price, highest price, lowest price, closing price, and trading volume; The data processing module is used to cache the acquired market data into memory and update it in real time. It uses a double-ended queue structure to cache the latest 6,000 K-line data. Technical indicator calculation module, used to calculate Kalman filter, Bayesian inference, E40 resonance, E41 resonance, Bollinger band and other technical indicators, and generate market analysis results based on these indicators; The trading signal generation module is used to generate trading signals based on the calculation results of multiple technical indicators to determine whether to enter the market; The decision-making execution module is used to automatically execute buy and sell instructions according to trading signals and complete transactions through the trading platform API interface; The risk control and position management module is used to dynamically adjust positions based on market volatility and set stop-loss and take-profit rules; The strategy optimization and backtesting module is used to backtest trading strategies using historical data, evaluate their performance and optimize technical indicator parameters.

[0008] Preferably, the technical indicator calculation module includes: Kalman filter module, which is used to predict future price trends based on historical market data and reduce the impact of price fluctuations on decision-making; Bayesian inference module, which is used to update the prior probability of market parameters based on historical data to dynamically adjust trading strategies; E40 resonance module, which is used to calculate the exponentially weighted moving average with a period of 40 minutes and generate resonance signals to determine the timing of buying and selling; E41 resonance module, which is used to calculate the exponentially weighted moving average with a period of 41 minutes and generate resonance signals to further confirm the buying and selling opportunities; The Bollinger Band calculation module is used to calculate the upper and lower bands of the Bollinger Bands based on price volatility and determine whether the market breaks through the upper and lower bands.

[0009] Preferably, the transaction signal generation module generates a transaction signal based on the following steps: Determine the market trend based on the signals of E40 resonance and E41 resonance; Determine whether the market breaks through the upper and lower tracks of the Bollinger Bands. If so, a trading signal is generated; Combine the market forecast value after Kalman filtering and the market parameters updated by Bayesian inference to generate buy or sell instructions.

[0010] Preferably, the risk control and position management module includes: The volatility assessment module is used to calculate market volatility based on historical market price data and dynamically adjust trading positions according to volatility; The stop loss and take profit module is used to set fixed stop loss and take profit points, where the stop loss point is 30 points and the take profit point is 41 points; The slippage control module is used to simulate the impact of slippage according to market volatility and make adjustments when executing transactions to reduce the impact of slippage on transaction results.

[0011] Preferably, the decision execution module executes the transaction through the following steps: Determine whether to enter the market based on the generated trading signal. If the entry conditions are met, issue a buy order through the API interface; When the stop loss or take profit conditions are met, the position is automatically closed and the sell order is executed; Dynamically adjust positions to ensure that the risk of each transaction is within a controllable range.

[0012] Preferably, the strategy optimization and backtesting module includes: Backtesting module, which is used to simulate the effect of backtesting trading strategies based on historical market data and evaluate the performance of strategies under different market conditions; The genetic algorithm optimization module is used to optimize the parameter configuration of technical indicators through genetic algorithms to improve the profitability of trading strategies.

[0013] The quantitative trading method of Hang Seng Index futures based on multi-indicator analysis includes the following steps: S1. Obtain real-time market data from the Hang Seng Index Futures Trading Platform and cache the data into memory; S2. Calculate multiple technical indicators, including Kalman filter, Bayesian inference, E40 resonance, E41 resonance, and Bollinger Bands; S3. Based on the calculation results of multiple technical indicators, determine the entry and exit timing of the market and generate corresponding buy and sell signals; S4. Automatically execute trading operations through the API interface based on the generated trading signals; S5. Adjust positions according to market volatility and set stop loss and take profit points; S6. Backtest the trading strategy and optimize the strategy parameters through genetic algorithm.

[0014] Preferably, the step of calculating the technical indicators includes: S21. Smooth historical data through Kalman filtering to predict future price changes in the market; S22. Use Bayesian inference method to update market parameters and update the prior probability of market status through historical data; S23. Calculate the exponentially weighted moving average with a period of 40 minutes and 41 minutes and determine the resonance signal; S24. Calculate the upper and lower Bollinger Bands based on the volatility of market prices, and determine whether the price breaks through the upper and lower bands.

[0015] Preferably, the transaction signal generating step comprises: S41. When the E40 resonance and E41 resonance signals are consistent and the market price breaks through the Bollinger Bands, a buy or sell signal is generated; S42. Determine whether to confirm the buy signal based on the market trend predicted by the Kalman filter; S43. Further confirm whether to execute buying or selling based on the updated market parameters after Bayesian inference.

[0016] Preferably, the risk control and position management steps include: Calculate volatility based on historical market price data and adjust positions based on volatility; Set fixed stop loss and take profit points, the stop loss point is 30 points, and the take profit point is 41 points; Simulate slippage during volatile market conditions and adjust the price at which trades are executed to reduce the risk of slippage.

[0017] The present invention provides a Hang Seng Index futures quantitative trading system and method based on multi-index analysis. It has the following beneficial effects: 1. The present invention adopts a quantitative trading system based on multi-indicator analysis, combined with multiple technical indicators such as Kalman filtering, Bayesian inference, E40 and E41 resonance, Bollinger Bands, etc., to achieve the technical effect of accurately capturing market trends, volatility and reversal signals. Compared with the strategy of single indicator analysis in the prior art, the present invention can comprehensively analyze market dynamics from multiple angles, solve the problem of signal distortion that may be caused by a single technical indicator, and thus significantly improve the accuracy and reliability of trading decisions.

[0018] 2. The present invention can execute trading instructions in real time and dynamically adjust positions and set stop-loss and take-profit points according to market fluctuations through automated decision-making execution and risk control mechanisms. This technical solution solves the delay and error problems that may be caused by manual operations in the prior art, greatly improves the speed and accuracy of transaction execution, reduces the risks caused by human decision-making, and enables the system to maintain stability and efficiency in high-frequency trading.

[0019] 3. The present invention automatically optimizes the trading strategy by combining the backtesting and genetic algorithm optimization modules, and continuously adjusts the parameters of the technical indicators, thereby improving the adaptability of the strategy. Compared with the traditional fixed parameter trading strategy, the present invention can optimize the parameter configuration under different market conditions and adjust the trading strategy in real time, solving the limitation of the static strategy in the prior art that it is difficult to adapt to market changes, and enhancing the profitability of the system in a dynamic market environment.

[0020] 4. The present invention adopts a double-ended queue cache mechanism in the data acquisition and processing module to ensure the real-time and high efficiency of the data. This technical solution not only optimizes the way of data storage and updating, but also solves the problems of data delay and expiration in the prior art. By acquiring the latest market data in real time and updating the cache efficiently, the present invention can ensure that the subsequent technical indicator calculation and trading signal generation are based on the latest market dynamics, ensuring the system's response speed and data processing capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a framework diagram of the quantitative trading system of the present invention; Figure 2 It is a technical indicator analysis flow chart of the present invention; Figure 3 A diagram of the data processing and caching mechanism of the present invention; Figure 4 This is a flow chart of the automated transaction execution and risk control of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] With the rapid development of financial markets, especially in complex trading environments such as high-frequency trading, traditional single technical indicator analysis methods can no longer meet the needs of rapidly changing markets. Market trends, volatility, and reversal signals change extremely quickly, and decision-making systems that rely on a single indicator often have lags and are difficult to respond to market changes in a timely manner. At the same time, existing automated trading systems often rely on fixed strategies and parameters and lack the ability to dynamically adapt to changes in market conditions. In addition, data delays and inadequate caching mechanisms often lead to missed optimal trading opportunities, affecting the overall performance and profitability of the system.

[0024] In order to solve these problems, the present invention proposes a quantitative trading system based on comprehensive analysis of multiple technical indicators, automated decision execution and optimization algorithm. The system can not only effectively improve the accuracy, flexibility and timeliness of trading decisions, but also adjust strategies in real time under different market environments, thereby improving overall trading performance and profitability.

[0025] Please refer to the attached Figure 1 The embodiment of the present invention provides a Hang Seng Index futures quantitative trading system based on multi-indicator analysis, including a data acquisition module, a data processing module, a technical indicator calculation module, a trading signal generation module, a decision execution module, a risk control and position management module, and a strategy optimization and backtesting module. The following is a detailed description of each module.

[0026] Data acquisition module: Please refer to the attached Figure 3 In the present invention, the data acquisition module is the starting link of the quantitative trading system, and its main function is to obtain real-time market data from the Hang Seng Index Futures Trading Platform. These data include but are not limited to 1-minute K-line data (opening price, highest price, lowest price, closing price and trading volume), as well as other market data that may be needed (such as buy and sell order depth, trading volume, etc.). The data acquisition module is the basis of the entire trading system, and its performance directly affects the effects of subsequent technical indicator calculations, trading signal generation and decision execution modules.

[0027] In this embodiment, the data acquisition module connects to the API interface of the Hang Seng Index Futures Trading Platform to obtain market data from the exchange in real time. The system sends a request every second to obtain the latest market data. Each time the data obtained from the platform includes 1 minute of K-line data, specifically including: opening price ( )、Highest Price( )、Minimum Price( )、Closing Price( ) and volume ( ). These data can be transmitted to the data processing module in real time for subsequent processing and analysis.

[0028] Specifically, the operation process of the data acquisition module is as follows: Send a request to the trading platform API interface to obtain the latest market data. The API request includes parameters such as timestamp, market identifier, data type, etc.

[0029] The acquired market data (i.e. K-line data and other relevant market data) is transmitted to the data processing module.

[0030] Based on the acquired K-line data and market information, the system formats each data point to ensure that the data can be effectively read by subsequent modules.

[0031] As an option, the data acquisition module not only obtains the 1-minute K-line data of the Hang Seng Index futures, but also selects data of other frequencies according to market demand, such as K-line data of 5-minute, 10-minute and other periods, or selects other types of market data, such as buying and selling depth data, trading volume data, etc., so that the system can select the most appropriate data for processing in different trading strategies.

[0032] In general, to ensure the real-time and completeness of data, the data acquisition module will pull data once a second to ensure that the system can provide the latest market dynamics at all times. By connecting to the exchange's API interface, the system can monitor market changes in real time and adjust trading strategies in a timely manner based on the acquired data.

[0033] In a possible implementation, the data acquisition module can also use a data cache mechanism to avoid data loss or delay. The system stores the received data in memory and sorts them by timestamp to ensure that the latest data is always at the forefront of the cache for use by subsequent modules.

[0034] In summary, the data acquisition module ensures that the quantitative trading system can analyze and make decisions based on the latest market data through efficient data capture and real-time update mechanisms. This module provides real-time and accurate market data support for the entire trading system and is the basis for the stable operation of the system.

[0035] Data processing module: Please refer to the attached Figure 3 ,After the data acquisition module obtains market data from the Hang Seng Index Futures Trading Platform, the ,data processing module receives the data and further processes and formats it for ,subsequent technical indicator calculation, trading signal generation and other modules.,The main task of the data processing module is to ensure the ,quality and consistency of the data and provide clean data for ,processing by various analysis modules.

[0036] In this embodiment, the data processing module uses a double-ended queue (deque) structure to cache real-time market data. The double-ended queue structure has efficient insertion and deletion performance, can quickly store newly acquired market data, and automatically discard the oldest data. In this way, the system ensures that the latest 6,000 K-line data are used in each calculation to ensure the timeliness and continuity of the data.

[0037] Specifically, the workflow of the data processing module includes the following steps: Receiving data: The data acquisition module passes the market data (including opening price, highest price, lowest price, closing price and trading volume) obtained from the trading platform to the data processing module. The data processing module inserts the data into the tail of the double-ended queue according to the timestamp of the data.

[0038] Cache update: The data processing module periodically checks the queue length. If the queue length exceeds the preset maximum value (for example, 6,000 K-line data), the oldest data is deleted. This ensures that the data cache always keeps the latest market information and avoids excessive memory usage.

[0039] Data formatting: When receiving new data, the data processing module will also format each piece of data to ensure that the data meets the input requirements of subsequent modules. For example, the data's timestamp, price, and transaction volume will be standardized to ensure the consistency and operability of each piece of data.

[0040] Data storage and management: All cached data will be stored in chronological order and updated according to the latest data. Every time new data is added, the queue status will be updated in the data processing module to ensure the orderliness and real-time nature of the data.

[0041] As an option, the data processing module can also perform preliminary cleaning of the data during storage to remove outliers or erroneous data. For example, when market data is missing or erroneous, the system can automatically perform interpolation or discard the data point to ensure that the data quality meets the analysis requirements.

[0042] In general, the data processing module not only processes 1-minute K-line data, but also processes data of different periods. If the system requires data with a higher frequency, the data processing module can adjust the cache capacity and update frequency to meet the needs of data with different periods. For example, if the system needs to process 5-minute K-line data, the cache capacity and data update frequency will be adjusted accordingly to meet the computing needs of a longer time period.

[0043] In a possible implementation, the data processing module can also add a time synchronization mechanism during the data storage and processing process. Since market data may have network delays or timestamp deviations, the system can synchronize the data to ensure that the timestamps of all data are consistent, thereby ensuring the accuracy and synchronization of data processing.

[0044] In summary, the data processing module ensures efficient storage and management of real-time data through the cache mechanism of the double-ended queue structure. This module not only provides continuous data updates, but also ensures the accuracy and consistency of the data, providing a reliable data foundation for subsequent technical indicator calculations and trading signal generation modules.

[0045] Technical indicator calculation module: Please refer to the attached Figure 2In the present invention, the technical indicator calculation module is the core part of the quantitative trading system. Its main function is to calculate multiple technical indicators based on market data and use these indicators to generate trading signals. The real-time market data provided by the data acquisition and processing module will enter the technical indicator calculation module for further analysis after being cached. The calculation process of this module is automated and is based on a variety of mathematical models and statistical methods for in-depth analysis to reveal the potential trends and price fluctuations of the market.

[0046] In this embodiment, the technical indicator calculation module mainly includes the Kalman filter module, the Bayesian inference module, the E40 and E41 resonance module, and the Bollinger band calculation module. Each module performs different calculation tasks independently, while cooperating with each other to provide support for trading decisions. Specifically, these modules process market data not only considering the current price, but also predicting future price changes based on historical data, providing an accurate reference for the trading signal generation module.

[0047] Kalman filter module Specifically, the Kalman filter module is used to smooth market data, reduce noise interference, and provide potential trend directions for trading decisions by predicting historical data. Kalman filtering is calculated in two main steps: prediction and update. First, the system predicts the future trend of the market based on historical data, and then updates and corrects the prediction results based on actual market data. By continuously updating the predicted values, the Kalman filter module can effectively identify potential trends in price fluctuations.

[0048] During implementation, the Kalman filter module uses the following formula:

[0049] in: For the moment The predicted value of .

[0050] For the moment The actual observed value of (i.e. the actual price in the market).

[0051] is the Kalman gain, which controls the weight between the predicted value and the actual observed value.

[0052] is the observation matrix, usually the identity matrix, which represents the direct observation of the state.

[0053] In some embodiments, the Kalman filter module focuses on separating short-term market fluctuations from long-term trends, so that the system can more stably capture trend changes while ignoring short-term noise.

[0054] Bayesian Inference Module As an option, the Bayesian inference module provides a way for the system to dynamically adjust its strategy. The core idea of ​​Bayesian inference is to update the prior probability of the market based on historical data, so that the system can adjust itself in real time according to market changes. In this way, the system can not only respond to known market laws, but also respond flexibly to unknown market changes.

[0055] Bayesian inference is calculated using the following formula:

[0056] in: is the posterior probability, which means that given the known market data Next, parameters possibility.

[0057] is the likelihood function, which means that given the parameters When the market data Probability of occurrence.

[0058] is the prior probability, indicating that when there is no observed data, the parameter possibility.

[0059] is the marginal likelihood of the data, indicating that the observed data probability.

[0060] Specifically, Bayesian inference enables the system to update the parameters of the trading strategy based on real-time market data, such as the adjustment of stop loss and take profit points. Through the updated posterior probability, the system can make decisions that are more in line with the current market conditions in a dynamic market environment.

[0061] E40 and E41 Resonance Modules In some embodiments, the E40 and E41 resonance modules are used to capture market reversal signals. The E40 resonance module calculates the short-term trend of the market based on a 40-minute exponentially weighted moving average (EWMA); the E41 resonance module uses a 41-minute EWMA for similar calculations. By calculating the weighted moving average of these two periods, the system can detect the crossover of the short-term moving average and the long-term moving average, thereby determining whether the market has reversed.

[0062] The formula for E40 resonance calculation is:

[0063] in: is the smoothing factor, which determines the weight of the current price in the calculation.

[0064] For the moment market price.

[0065] For the moment The weighted average of the 40-minute periods.

[0066] The calculation formula of the E41 resonance module is similar to that of the E40 resonance module, but its period is 41 minutes:

[0067] Generally speaking, when the two resonance signals E40 and E41 cross at the same time, the system will generate a trading signal, indicating that the market may reverse. These reversal signals are usually very important decision-making basis in trading strategies.

[0068] Bollinger Bands calculation module As another option, the Bollinger Bands calculation module is used to assess market volatility. Bollinger Bands determine the upper and lower bands by calculating the standard deviation of market prices to determine whether the market is overly volatile. When the market price breaks through the upper and lower bands of the Bollinger Bands, it may mean that the market has entered an extreme state, thus generating a trading signal.

[0069] The Bollinger Bands calculation formula is as follows:

[0070] in: For the moment The moving average is usually the average value of a fixed period (such as 20 minutes).

[0071] is the standard deviation of the price, which indicates the amplitude of market price fluctuation.

[0072] It is a constant, usually 2, which represents the width of the Bollinger Band.

[0073] Specifically, the upper and lower bands of the Bollinger Bands are determined by market price volatility. If the market price approaches or breaks through the upper band, the system may consider the market to be overbought and generate a sell signal; if the market price approaches or breaks through the lower band, the system considers the market to be oversold and may generate a buy signal.

[0074] During the implementation process, the technical indicator calculation module provides rich market information for the system's trading decisions through the comprehensive calculation of multiple technical indicators such as Kalman filtering, Bayesian inference, E40 and E41 resonance, and Bollinger Bands. These indicators analyze market trends, volatility, and reversal signals from different perspectives, enabling the system to make efficient and accurate decisions in a complex and changing market environment. Through the collaborative work of these technical indicators, the quantitative trading system of the present invention can dynamically adapt to market changes and capture the best trading opportunities, thereby improving the success rate of transactions and reducing risks.

[0075] Trading signal generation module: In the quantitative trading system of the present invention, the trading signal generation module is the key link connecting the technical indicator calculation module and the decision execution module. Its function is to automatically generate buy or sell signals under the comprehensive analysis of multiple technical indicators. Trading signal generation not only depends on each independent technical indicator, but also needs to consider the mutual coordination between indicators and market dynamic changes. While ensuring the rationality of the trading strategy, this module maximizes the capture of trading opportunities in market fluctuations and provides an accurate basis for decision execution.

[0076] In this embodiment, the trading signal generation module first receives the output results of the technical indicator calculation module, which include the calculated values ​​of technical indicators such as Kalman filtering, Bayesian inference, E40 and E41 resonance, and Bollinger Bands. By comprehensively analyzing these technical indicators, the system automatically generates trading signals and decides whether to perform a buy or sell operation.

[0077] Specifically, when generating a trading signal, the system will first check whether there are multiple technical indicators giving consistent signals. If multiple indicators (such as E40 resonance, E41 resonance and Bollinger Bands) send out buy signals at the same time, and the results of Kalman filtering and Bayesian inference also support the upward trend of the market, the system will believe that the market is at a good time to buy, and then generate a buy signal.

[0078] As an option, if multiple technical indicators give contradictory signals, the system will process them according to the preset priority. For example, the system can determine the short-term market trend based on the signals of E40 and E41 resonance, and then confirm the final trading signal based on the long-term trend predicted by the Kalman filter. If the trend predicted by the Kalman filter is consistent with the signal of E40 resonance, the system will prioritize the signal and ignore other inconsistent signals.

[0079] Generally speaking, when the breakthrough signals of the upper and lower rails of the Bollinger Bands are combined with the signals of other technical indicators, the reliability of the signals will be further enhanced. For example, when the market price breaks through the upper rail of the Bollinger Bands, it means that the market may have been overbought. At this time, if the predicted values ​​of Kalman filtering and Bayesian inference also support the market going down, the system will automatically generate a sell signal.

[0080] Specifically, the workflow of the trading signal generation module can be described as follows: E40 resonance and E41 resonance: By calculating the exponentially weighted moving average with a period of 40 minutes and 41 minutes, the system determines the short-term reversal signal of the market. If the E40 and E41 resonances give a buy or sell signal at the same time, the system will further verify the trend of other indicators to ensure the reliability of the signal.

[0081] Kalman filtering and Bayesian inference: Kalman filtering is used to smooth market data and reduce noise interference, while Bayesian inference updates the prior probability of the market in real time. Through these tools, the system can predict the long-term trend of the market and help determine whether to continue holding or closing positions.

[0082] Bollinger Band Breakout: The upper and lower bands of the Bollinger Bands are used to determine the volatility of the market. When the market price breaks through the upper band of the Bollinger Bands, the system will generate a sell signal. Conversely, when the price breaks through the lower band, the system will generate a buy signal.

[0083] Comprehensive judgment: If the signals of the above-mentioned multiple technical indicators are consistent and in line with the current market trend, the system will generate specific operation signals for buying or selling.

[0084] In some embodiments, the system can also dynamically adjust the weight of trading signals according to different stages of the market. For example, when the market is in a range of fluctuations, the signal of the Bollinger Bands may have a higher weight; while in a trending market, the signal of the resonance of E40 and E41 may be given priority.

[0085] As another possible implementation method, the system can also use machine learning algorithms to optimize the generation process of trading signals based on the historical market data and current market conditions. For example, the system can learn which technical indicator combinations perform better under specific market conditions based on historical trading data, thereby dynamically adjusting and optimizing the generation rules of trading signals.

[0086] In the present invention, the generation of trading signals depends on multiple technical indicators. The following is a detailed description of the calculation formulas and related parameters of the aforementioned technical indicators to ensure that the parameter definitions in the technical implementation are complete and without omissions.

[0087] E40 Resonance and E41 Resonance: Exponentially Weighted Moving Average formula (EWMA) used when calculating short-term trend reversal signals.

[0088] Likewise, the E41 Resonance module uses a similar formula, but with the period adjusted to 41 minutes.

[0089] Kalman Filter: The Kalman filter adjusts the estimate of market trends through prediction and update steps.

[0090] Bayesian inference: used to dynamically adjust prior probabilities of market states.

[0091] Bollinger Bands: The upper and lower bands of the Bollinger Bands are calculated using the standard deviation of market prices.

[0092] By integrating the trading signals generated by these technical indicators, the system can respond quickly in the real-time market and automatically make decisions to buy or sell according to the market status. The combination of all these technical indicators enables the present invention to operate stably in a complex market environment and effectively capture potential trading opportunities.

[0093] Decision execution module: In the present invention, the decision execution module is a key component of the quantitative trading system, which is responsible for automatically executing trading operations according to the buy or sell signals output by the trading signal generation module. The goal of this module is to convert the analysis results into actual market operations, ensure that the system can buy or sell futures contracts at the right time, and realize the automatic execution of trading strategies. In this process, the decision execution module needs to work closely with the technical indicator calculation module and the trading signal generation module to ensure the real-time and accuracy of trading decisions.

[0094] In this embodiment, after the decision execution module receives the buy or sell signal from the transaction signal generation module, it will first confirm the signal. This process includes the verification of multiple technical indicator signals and the judgment of the market status. Only when multiple technical indicators are consistent and meet the current market conditions will the system execute the corresponding transaction operation. In addition, the system will dynamically adjust the execution strategy based on the real-time data of the market to ensure the best trading time.

[0095] Specifically, the decision execution module decides whether to enter the market based on the signal of the trading signal generation module. After confirming that the trading signal is valid, the module will call the API interface of the trading platform to issue a buy or sell order. The specific operation process is as follows: Signal confirmation: Once the trading signal generation module generates a buy or sell signal, the decision execution module will first confirm the signal. The basis for confirmation is whether multiple technical indicators are consistent. If multiple technical indicators (such as Kalman filter, E40 resonance, E41 resonance, Bollinger band, etc.) all support the market trend in the same direction, the system considers the trading signal valid and then executes the transaction.

[0096] Market status judgment: After the signal is confirmed, the decision-making execution module will also judge the timing of the signal execution based on the current market status. For example, the system will determine whether to execute the transaction immediately or wait for the market to further confirm the reliability of the signal based on factors such as market volatility and trading volume.

[0097] Execute a transaction: Once the transaction signal is confirmed, the system will send a buy or sell instruction to the trading platform through the API interface. This process is automated and does not require human intervention, ensuring that the transaction can be completed in the fastest time.

[0098] Stop loss and take profit settings: While executing trading instructions, the decision execution module will also set corresponding price limits according to the preset stop loss and take profit rules. The system will automatically set stop loss points and take profit points according to the opening price of the transaction, so that the position will be automatically closed when the market price reaches these points, ensuring that the transaction risk is controlled within an acceptable range. Generally, the stop loss point is set at 30 points and the take profit point is set at 41 points.

[0099] As an option, if the market price fluctuates greatly, the decision execution module can automatically adjust the stop loss and take profit points. For example, when the market price fluctuates greatly, the system can increase the stop loss point setting to avoid hitting the stop loss in the sharp fluctuations in the short term; similarly, when the market price fluctuates less and is stable, the take profit point can be reduced accordingly to maximize the profit.

[0100] In one possible implementation, the decision execution module can also be adjusted according to the market depth data. For example, when the market buying or selling is strong, the system may choose to enter or exit the market early, thereby reducing the impact of slippage and obtaining a better transaction price.

[0101] Generally, when executing a transaction, the decision execution module will also simulate and calculate the possible slippage. By estimating the market volatility and current trading volume, the system can predict the slippage that may occur during the actual transaction execution process and adjust the execution price when the transaction signal is generated. In this way, the system can not only ensure that the transaction is executed according to the predetermined signal, but also reduce the negative impact of slippage on the transaction effect.

[0102] Specifically, the operation process of the decision execution module can be further refined into the following steps: Triggering of trading signals: When the trading signal generation module sends a buy signal, the system first determines whether the current market price meets the entry conditions. If the market price meets the conditions for signal generation, the system will trigger a buy operation.

[0103] Market status judgment: When judging the entry time, the system will conduct a comprehensive analysis based on the current market depth, trading volume and market volatility. If the market is volatile and the trading volume increases, the system will give priority to executing the transaction to ensure that the market can be entered as quickly as possible.

[0104] Execute buy operation: After confirming that the buy signal is valid, the system will send a buy order to the trading platform and record the current opening price. At this time, the system will also set the corresponding stop loss and take profit points.

[0105] Automatic stop loss and take profit: When trading, the system will monitor the market price according to the set stop loss and take profit points. Once the market price reaches the stop loss or take profit point, the system will automatically close the position and generate a sell signal.

[0106] Closing operation: If the market price reaches the stop loss or take profit point, the system will execute a sell operation based on the closing signal to complete the entire transaction.

[0107] As another option, when market prices fluctuate greatly, the decision execution module can also make adjustments by simulating slippage. The system will predict possible slippage based on the current market depth, number of pending orders, and market volatility, and adjust the transaction execution price based on the predicted value. In this way, the system can reduce the transaction costs caused by slippage and ensure that the transaction price is closer to the expected level.

[0108] In the implementation of the present invention, the core of the decision-making execution module is to realize the automatic execution of trading instructions by connecting with the API interface of the trading platform. All trading operations, including buying, selling, stop loss and closing positions, are automatically executed by the system after receiving the trading signal without manual intervention. The system monitors the market status in real time, determines the best trading opportunity, and ensures the safety of transactions through a preset risk control mechanism.

[0109] To ensure smooth transaction execution, the decision execution module also considers the following factors: Real-time: Due to the high volatility of the market, the decision-making execution module must be sufficiently real-time to complete the transaction execution as quickly as possible after the signal is generated.

[0110] Risk control: While executing transactions, the system needs to monitor market prices in real time to ensure that the risks of transactions are within controllable range. The system's stop loss and stop profit mechanisms can effectively avoid excessive losses caused by market fluctuations.

[0111] Slippage control: In order to ensure the best execution price for the transaction, the system also needs to simulate the impact of slippage and fine-tune the execution price according to market conditions to reduce losses caused by slippage.

[0112] In summary, the decision-making execution module in the present invention provides the system with efficient and stable trading capabilities through automated transaction execution and precise risk control. The module can not only respond to market signals in real time and quickly execute buying and selling operations, but also minimize risks and ensure smooth transactions through stop-loss, stop-profit and slippage control mechanisms. Through this module, the quantitative trading system can flexibly respond to complex and changing market environments and bring stable returns to investors.

[0113] Risk control and position management module: Please refer to the attached Figure 4 In the quantitative trading system of the present invention, the risk control and position management module is one of the core modules to ensure the safe and robust execution of trading strategies. It is mainly responsible for dynamically adjusting the position size of transactions in the case of market fluctuations, and effectively controlling risks through stop-loss and stop-profit mechanisms. This module works closely with the decision-making execution module to automatically adjust the position size of each transaction by real-time monitoring of market volatility and combining the risk threshold set by the system to ensure maximum protection of capital security during the transaction process.

[0114] In this embodiment, the main function of the risk control and position management module is to dynamically adjust the position size according to the market volatility and the risk control rules of the system, and set the stop loss and take profit mechanism to ensure that the transaction risk is effectively controlled in an uncertain market environment. This module is not only responsible for adjusting the position, but also sets the stop loss point and take profit point in real time when executing the transaction, ensuring that the position can be closed in time when the market price fluctuates violently to prevent excessive losses.

[0115] Position Management Specifically, the position management part determines the current position size by calculating the volatility of the market. The system monitors the fluctuation of market prices in real time and adjusts the position according to the volatility. When the market fluctuates greatly, the system will reduce the position to reduce potential risks; when the market fluctuates less and the price trend is clear, the system can appropriately increase the position to increase the profit opportunity.

[0116] The core basis of position management is market volatility, which can be measured by calculating the standard deviation of market prices. The calculation formula is:

[0117] in: For the moment market price data.

[0118] is the average market price.

[0119] is the time period used to calculate volatility.

[0120] Generally speaking, the greater the volatility, the higher the market uncertainty, and the system will reduce positions to avoid excessive exposure to market risks. When volatility is low and the market is more stable, the system can appropriately increase positions to capture more market opportunities.

[0121] As an option, the system can determine the magnitude of position adjustments based on the set risk tolerance. For example, if the risk threshold set by the system is high, the magnitude of position adjustments will be large; if the risk threshold is low, the magnitude of position adjustments will be small, ensuring that the system trades at a lower risk.

[0122] In some embodiments, the risk control module also includes stop loss and stop profit settings. Stop loss and stop profit are to avoid huge losses caused by market fluctuations and automatically close positions when the market reaches a preset profit point. The system ensures that the risk of each transaction can be controlled within an acceptable range by setting stop loss points and stop profit points.

[0123] The setting of stop loss points is usually based on market volatility and risk control rules set by the user. In the present invention, the system usually sets the stop loss point to a fixed 30 price points and the take profit point to 41 price points. These values ​​can be dynamically adjusted according to actual market conditions and user needs. Specifically, the system automatically calculates the stop loss point and the take profit point based on the real-time market price, and automatically executes the closing operation when the price reaches these points.

[0124] For example, in a transaction, assuming the market price is , the system will calculate the stop loss price based on the preset stop loss point and take profit point and take profit price When the market price reaches these two points, the system will automatically close the position and generate a sell signal.

[0125] Specifically, the system monitors market price changes in real time. When the market price reaches the stop loss point or the take profit point, the system automatically sends a closing instruction to the trading platform. In this way, the system ensures that each transaction can be completed within the preset risk range, thereby avoiding losses caused by human misjudgment or delays.

[0126] As an option, in a highly volatile market, the system can automatically adjust the size of the stop loss and take profit points according to volatility. If the market volatility is high, the system will automatically expand the stop loss and take profit points to prevent frequent stop losses caused by short-term price fluctuations. When the market volatility is low, the system will reduce the stop loss and take profit points to ensure that profits can be quickly locked in when the market is more stable.

[0127] In another embodiment, the risk control and position management module also includes a slippage control mechanism. Slippage refers to the difference between the actual transaction price and the expected price caused by market price fluctuations. Slippage usually occurs when market liquidity is low or price fluctuations are large. In order to reduce the impact of slippage on transaction results, the system will simulate the possibility of slippage during the transaction execution process and fine-tune the order execution price based on this.

[0128] Specifically, the system predicts the size of the slippage based on the current volatility and depth information of the market. If the market price fluctuates greatly, the system may adjust the buy or sell price based on the predicted slippage to ensure that the actual transaction price is close to the expected value. For example, if the predicted slippage is 5 price points, the system may slightly adjust the buy price when generating a buy signal to ensure that the price is close to the best price in the market.

[0129] In one possible implementation, the slippage control mechanism can also be used in conjunction with market depth data. By obtaining real-time buy and sell order data, the system can assess the current market liquidity and adjust the transaction execution strategy based on the depth data. If the market liquidity is low, the system will reduce the transaction volume or postpone the transaction execution accordingly until the market liquidity recovers.

[0130] Through the risk control and position management module, the present invention can automatically adjust the position size of the transaction in the case of market fluctuations, set the stop loss and take profit points in real time, and ensure that each transaction is completed within the controllable risk range. The volatility calculation and slippage control mechanism further enhance the system's ability to cope with high volatility markets. In addition, the system can also dynamically adjust the risk management strategy according to the real-time status of the market, thereby effectively improving the stability and security of transactions.

[0131] Generally speaking, this module can automatically execute risk control and position management strategies without human intervention, thereby reducing the risk of human decision-making errors, ensuring that the system can operate stably in various market environments, maximizing profits and effectively avoiding losses.

[0132] Strategy optimization and backtesting module: Please refer to the attached Figure 4 In the present invention, the strategy optimization and backtesting module is an important part of the quantitative trading system, responsible for evaluating and optimizing the effectiveness of the trading strategy. The module backtests the trading strategy through historical market data, simulates the performance of the strategy under different market environments, and thus helps the system identify the optimal trading parameters and dynamically optimizes the strategy to adapt to changing market conditions. The module works closely with the technical indicator calculation module, the trading signal generation module, and the decision execution module to ensure that the system can be continuously adjusted and optimized in actual operation to obtain better trading results.

[0133] In this embodiment, the main function of the strategy optimization and backtesting module is to backtest the trading strategy through historical data, evaluate the effectiveness of the strategy, and use optimization algorithms (such as genetic algorithms) to adjust the parameters of technical indicators, thereby improving the profitability and adaptability of the strategy. The backtesting and optimization process is automated, and can continuously improve the performance of the trading strategy while continuously processing and analyzing data.

[0134] Backtesting Module Specifically, the backtesting module uses historical market data to simulate actual transactions to evaluate the performance of the selected strategy under different market environments. The backtesting process includes the following steps: Data preparation: The backtesting module first extracts necessary market data from historical data, including price, trading volume and other information. These data are used to simulate the conditions of actual transactions. Usually, the backtesting module will use historical K-line data and other relevant market data, such as depth data, trading volume, etc.

[0135] Strategy execution: According to the set trading strategy, the backtesting module will simulate the execution of transactions in different time periods. By using historical data, the backtesting module will gradually execute trading instructions, record the opening price, closing price, stop loss point and take profit point of each transaction, and calculate the profit of each transaction.

[0136] Result analysis: After the backtest is completed, the backtest module will output a detailed analysis of the trading results, including key indicators such as the profit and loss of each transaction, total return, maximum drawdown, and Sharpe ratio. These results help the system evaluate the effectiveness and risk level of the strategy. Through these analyses, users can understand the performance of the strategy in the historical market environment and make corresponding adjustments.

[0137] Generally, the backtesting module will compare multiple different strategies and select the best performing strategy for optimization and implementation. The backtesting results can help identify the most promising trading strategies, thereby increasing the success rate of subsequent actual transactions.

[0138] Optimization Module As an option, after the backtest is completed, the strategy optimization module will further analyze the backtest results and find the optimal parameter configuration by adjusting the technical indicator parameters in the trading strategy. This process is usually achieved through optimization methods such as genetic algorithms. The core of genetic algorithms is to simulate the process of natural selection and find the best trading parameters through multiple generations of evolution and adaptive selection.

[0139] The optimization process of genetic algorithm can be summarized as the following steps: Initialization population: The system initializes the strategy population by randomly generating a certain number of trading strategy parameter combinations. These initial parameter combinations will be used to perform backtesting and evaluate their performance.

[0140] Select individuals with strong adaptability: Based on the backtest results, each individual (i.e., strategy parameter combination) will receive a fitness score, which indicates the performance of the combination in historical data. Individuals with high fitness will be selected as "parents" to generate the next generation.

[0141] Crossover and mutation: The selected parents generate new individuals (strategy parameter combinations) through crossover and mutation. The crossover operation simulates gene exchange in genetics, and the mutation operation randomly adjusts certain parameters to increase diversity.

[0142] Select the best individual: Through multiple generations of evolution, the genetic algorithm will continuously improve the strategy parameters and eventually find the optimal parameter combination to optimize the trading strategy. The best strategy parameters finally selected will be applied in actual trading.

[0143] In one possible implementation, the optimization module does not only rely on genetic algorithms, but can also be combined with other optimization algorithms, such as particle swarm optimization (PSO) or simulated annealing algorithms, to flexibly select the most appropriate optimization method according to different market conditions.

[0144] Specifically, the advantage of using the backtesting and optimization modules together lies in its automation and efficiency. Through backtesting, the system can quickly evaluate the performance of different strategies in the historical market, and through optimization, the system can automatically adjust the strategy parameters to improve the profitability and market adaptability of the strategy. The combination of the two ensures that the trading strategy can operate effectively under different market conditions and improves the success rate of the strategy in actual operation.

[0145] For example, suppose the backtest results show that in a certain market cycle, the existing stop loss and take profit settings are not ideal, resulting in some potential losses. At this time, the optimization module automatically adjusts the parameters of the stop loss and take profit points through genetic algorithms, and verifies the effect of the improved strategy again through backtesting, and finally selects the best strategy for actual trading.

[0146] In the backtesting and optimization process, the core formulas and parameters involved include: Backtest results evaluation indicators: Total Return: The total profit and loss of all trades, indicating the overall profitability of the strategy.

[0147] Maximum Drawdown: The maximum loss of account funds from the highest point to the lowest point during the drawdown period, which indicates the risk level of the strategy.

[0148] Sharpe ratio: used to measure the risk-adjusted return of a strategy, usually calculated as:

[0149] Among them, the average return represents the average profit of the strategy, and the standard deviation of the return represents the volatility of the strategy.

[0150] Genetic algorithm optimization parameters: Population Size: The number of strategy parameter combinations in each generation of the genetic algorithm.

[0151] Crossover Rate: The probability that a parent strategy parameter combination crosses to generate a child strategy parameter combination.

[0152] Mutation Rate: The probability of performing mutation operations on the offspring strategy parameter combination.

[0153] These indicators and parameters help the backtesting and optimization module to evaluate and improve trading strategies to achieve the best trading performance.

[0154] In general, the strategy optimization and backtesting modules play a vital role in the quantitative trading system of the present invention. The backtesting module provides historical data verification to ensure the rationality and effectiveness of the selected trading strategy, while the optimization module continuously improves the performance of the strategy through intelligent optimization methods such as genetic algorithms to ensure good profitability in a complex and changing market environment. Through the combination of backtesting and optimization modules, the system can automatically adjust the trading strategy to obtain better returns and reduce risks in actual transactions.

[0155] As part of this application, the present invention also provides a Hang Seng Index futures quantitative trading method based on multi-index analysis, comprising the following steps: S1. Obtain real-time market data from the Hang Seng Index Futures Trading Platform and cache the data into memory; S2. Calculate multiple technical indicators, including Kalman filter, Bayesian inference, E40 resonance, E41 resonance, and Bollinger Bands; The steps to calculate a technical indicator include: S21. Smooth historical data through Kalman filtering to predict future price changes in the market; S22. Use Bayesian inference method to update market parameters and update the prior probability of market status through historical data; S23. Calculate the exponentially weighted moving average with a period of 40 minutes and 41 minutes and determine the resonance signal; S24. Calculate the upper and lower Bollinger Bands based on the volatility of market prices, and determine whether the price breaks through the upper and lower bands.

[0156] S3. Based on the calculation results of multiple technical indicators, determine the entry and exit timing of the market and generate corresponding buy and sell signals; S4. Automatically execute trading operations through the API interface based on the generated trading signals; The trading signal generation steps include: S41. When the E40 resonance and E41 resonance signals are consistent and the market price breaks through the Bollinger Bands, a buy or sell signal is generated; S42. Determine whether to confirm the buy signal based on the market trend predicted by the Kalman filter; S43. Further confirm whether to execute buying or selling based on the updated market parameters after Bayesian inference.

[0157] S5. Adjust positions according to market volatility and set stop loss and take profit points; Risk control and position management steps include: Calculate volatility based on historical market price data and adjust positions based on volatility; Set fixed stop loss and take profit points, the stop loss point is 30 points, and the take profit point is 41 points; Simulate slippage during volatile market conditions and adjust the price at which trades are executed to reduce the risk of slippage.

[0158] S6. Backtest the trading strategy and optimize the strategy parameters through genetic algorithm.

[0159] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The Hang Seng Index futures quantitative trading system based on multi-index analysis is characterized by: include, Data acquisition module, used to obtain real-time market data from the Hang Seng Index Futures Trading Platform, including 1-minute K-line data, opening price, highest price, lowest price, closing price, and trading volume; The data processing module is used to cache the acquired market data into memory and update it in real time. It uses a double-ended queue structure to cache the latest 6,000 K-line data. Technical indicator calculation module, used to calculate Kalman filter, Bayesian inference, E40 resonance, E41 resonance, Bollinger band and other technical indicators, and generate market analysis results based on these indicators; The trading signal generation module is used to generate trading signals based on the calculation results of multiple technical indicators to determine whether to enter the market; The decision-making execution module is used to automatically execute buy and sell instructions according to trading signals and complete transactions through the trading platform API interface; The risk control and position management module is used to dynamically adjust positions based on market volatility and set stop-loss and take-profit rules; The strategy optimization and backtesting module is used to backtest trading strategies using historical data, evaluate their performance and optimize technical indicator parameters.

2. The Hang Seng Index futures quantitative trading system based on multi-index analysis according to claim 1 is characterized in that: The technical indicator calculation module includes: Kalman filter module, which is used to predict future price trends based on historical market data and reduce the impact of price fluctuations on decision-making; Bayesian inference module, which is used to update the prior probability of market parameters based on historical data to dynamically adjust trading strategies; E40 resonance module, which is used to calculate the exponentially weighted moving average with a period of 40 minutes and generate resonance signals to determine the timing of buying and selling; E41 resonance module, which is used to calculate the exponentially weighted moving average with a period of 41 minutes and generate resonance signals to further confirm the buying and selling opportunities; The Bollinger Band calculation module is used to calculate the upper and lower bands of the Bollinger Bands based on price volatility and determine whether the market breaks through the upper and lower bands.

3. The Hang Seng Index futures quantitative trading system based on multi-index analysis according to claim 1 is characterized in that: The trading signal generation module generates a trading signal based on the following steps: Determine the market trend based on the signals of E40 resonance and E41 resonance; Determine whether the market breaks through the upper and lower tracks of the Bollinger Bands. If so, a trading signal is generated; Combine the market forecast value after Kalman filtering and the market parameters updated by Bayesian inference to generate buy or sell instructions.

4. The Hang Seng Index futures quantitative trading system based on multi-index analysis according to claim 1 is characterized in that: The risk control and position management module includes: The volatility assessment module is used to calculate market volatility based on historical market price data and dynamically adjust trading positions according to volatility; The stop loss and take profit module is used to set fixed stop loss and take profit points, where the stop loss point is 30 points and the take profit point is 41 points; The slippage control module is used to simulate the impact of slippage according to market volatility and make adjustments when executing transactions to reduce the impact of slippage on transaction results.

5. The Hang Seng Index futures quantitative trading system based on multi-index analysis according to claim 1 is characterized in that: The decision execution module executes the transaction through the following steps: Determine whether to enter the market based on the generated trading signal. If the entry conditions are met, issue a buy order through the API interface; When the stop loss or take profit conditions are met, the position is automatically closed and the sell order is executed; Dynamically adjust positions to ensure that the risk of each transaction is within a controllable range.

6. The Hang Seng Index futures quantitative trading system based on multi-index analysis according to claim 1 is characterized in that: The strategy optimization and backtesting module includes: Backtesting module, which is used to simulate the effect of backtesting trading strategies based on historical market data and evaluate the performance of strategies under different market conditions; The genetic algorithm optimization module is used to optimize the parameter configuration of technical indicators through genetic algorithms to improve the profitability of trading strategies.

7. The Hang Seng Index futures quantitative trading method based on multi-index analysis is characterized by: The Hang Seng Index futures quantitative trading system based on multi-index analysis applied to any one of claims 1 to 6 comprises the following steps: S1. Obtain real-time market data from the Hang Seng Index Futures Trading Platform and cache the data into memory; S2. Calculate multiple technical indicators, including Kalman filter, Bayesian inference, E40 resonance, E41 resonance, and Bollinger Bands; S3. Based on the calculation results of multiple technical indicators, determine the entry and exit timing of the market and generate corresponding buy and sell signals; S4. Automatically execute trading operations through the API interface based on the generated trading signals; S5. Adjust positions according to market volatility and set stop loss and take profit points; S6. Backtest the trading strategy and optimize the strategy parameters through genetic algorithm.

8. The Hang Seng Index futures quantitative trading method based on multi-index analysis according to claim 7 is characterized in that: The steps of calculating the technical indicators include: S21. Smooth historical data through Kalman filtering to predict future price changes in the market; S22. Use Bayesian inference method to update market parameters and update the prior probability of market status through historical data; S23. Calculate the exponentially weighted moving average with a period of 40 minutes and 41 minutes and determine the resonance signal; S24. Calculate the upper and lower Bollinger Bands based on the volatility of market prices, and determine whether the price breaks through the upper and lower bands.

9. The Hang Seng Index futures quantitative trading method based on multi-index analysis according to claim 7, characterized in that: The transaction signal generating step comprises: S41. When the E40 resonance and E41 resonance signals are consistent and the market price breaks through the Bollinger Bands, a buy or sell signal is generated; S42. Determine whether to confirm the buy signal based on the market trend predicted by the Kalman filter; S43. Further confirm whether to execute buying or selling based on the updated market parameters after Bayesian inference.

10. The Hang Seng Index futures quantitative trading method based on multi-index analysis according to claim 7, characterized in that: Risk control and position management steps include: Calculate volatility based on historical market price data and adjust positions based on volatility; Set fixed stop loss and take profit points, the stop loss point is 30 points, and the take profit point is 41 points; Simulate slippage during volatile market conditions and adjust the price at which trades are executed to reduce the risk of slippage.

Images