Transaction execution control processing system based on transaction strategy
By introducing machine learning layer and trading strategy layer into the transaction execution control processing system, building a transaction prediction model and automatically executing trading signals, the problem of decision-making lag and insufficient adaptability in traditional systems when processing complex market data is solved, and the more Accurate and efficient transaction execution.
Patent Information
- Application Number
- CN202410994217.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-26
AI Technical Summary
When traditional transaction execution control processing systems process complex and changeable market data, it is difficult for manual analysis and judgment to accurately capture market dynamics, and automated transactions based on simple preset rules are difficult to adapt to complex market environments, resulting in decision-making lag or errors.
A transaction execution control processing system based on trading strategies is designed, including data collection layer, machine learning layer, transaction strategy layer, transaction execution layer, risk control layer and user interaction layer. Machine learning algorithms are used to build transaction prediction models and generate transactions. Signals and automatically perform buying and selling operations through the transaction execution layer to monitor and respond to risks in real time.
It improves the accuracy and efficiency of transaction decisions, reduces human intervention and misoperation, can respond more quickly to market changes, and ensures that transaction risks are within a controllable range.
Smart Images

Figure CN119027233A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of cross-trading execution, and in particular to a trading execution control processing system based on trading strategies. Background Art
[0002] In traditional transaction execution control processing systems, transaction decisions mainly rely on manual analysis and judgment, or automated transactions based on simple preset rules. These methods have obvious defects when processing complex and changeable market data. First, although manual analysis and judgment have certain flexibility and intuitiveness, they are often difficult to accurately capture market dynamics due to human cognitive ability and information processing speed. Especially when faced with large amounts of rapidly changing market data, manual decision-making is prone to lag and miss good trading opportunities. Secondly, although automated trading based on simple preset rules can improve transaction efficiency to a certain extent, these rules are often too simple and difficult to adapt to complex and changeable market environments. When unexpected changes occur in the market, such systems may not be able to respond effectively, and may even lead to wrong trading decisions. Therefore, we propose a transaction execution control processing system based on trading strategies. Summary of the invention
[0003] The purpose of the present invention is to provide a transaction execution control processing system based on transaction strategies to solve the problems raised in the above background technology.
[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a trading execution control processing system based on trading strategies, comprising a data collection layer, a machine learning layer, a trading strategy layer, a trading execution layer, a risk control layer and a user interaction layer;
[0005] The data collection layer includes a data acquisition module, a data cleaning module and a data preprocessing module;
[0006] The data acquisition module is used to capture the market data of local currency transactions from multiple data sources in real time; the data cleaning module is responsible for cleaning the captured data; the data preprocessing module is used to format and standardize the data;
[0007] The machine learning layer includes a data set partitioning module, a model selection and construction module, a model training module, and a model evaluation and tuning module;
[0008] The data set division module is used to divide the pre-processed historical data into a training set, a validation set and a test set; the model selection and construction module uses a machine learning algorithm to build a transaction prediction model, and the transaction prediction model is used to predict transaction signals in real time based on current market data; the model training module uses the training set data to train the transaction prediction model; the model evaluation and tuning module uses the validation set and the test set to evaluate the performance of the transaction prediction model;
[0009] The trading strategy layer includes a strategy logic definition module, a strategy signal generation module and a strategy verification module;
[0010] The strategy logic definition module defines the core logic and rules of the trading strategy according to business needs; the strategy signal generation module generates trading signals based on the trained machine learning model and current market data; the strategy verification module uses historical data and trading signals generated by the trading prediction model to backtest and verify the defined trading strategy, and transmits the verified favorable trading signals to the trading execution layer;
[0011] The transaction execution layer includes a transaction instruction generation module, a transaction instruction execution module and a transaction record management module;
[0012] The trading instruction generation module generates specific trading instructions according to the trading signals output by the trading strategy layer; the trading instruction execution module is responsible for connecting with the exchange interface to execute buying and selling operations; the trading record management module records the detailed information of each transaction;
[0013] The risk control layer includes a risk indicator calculation module, a risk assessment module and a risk response measure module;
[0014] The risk indicator calculation module calculates and monitors various risk indicators; the risk assessment module assesses potential risks based on market data and transaction conditions; the risk response module formulates corresponding measures based on the risk assessment results;
[0015] The user interaction layer includes a real-time market information display module, a trading strategy adjustment module, a trading status monitoring module and a historical trading record query module;
[0016] The real-time market information display module provides the latest market information to users; the trading strategy adjustment module allows users to adjust trading strategy parameters according to actual conditions; the trading status monitoring module displays the trading status in real time; and the historical trading record query module provides a historical trading record query function.
[0017] Preferably, the model selection and construction module uses a linear regression algorithm to construct a transaction prediction model. The linear regression algorithm uses the following formula to describe the relationship between the independent variable and the dependent variable:
[0018] Y=β0+β1X1+β2X2+…+β n X n +ε
[0019] Among them, Y is the dependent variable, representing the predicted trading signal, which includes the buying and selling direction, the price and quantity of the order; X1, X2, …, Xn is the independent variable, representing the input market data, including market price data, trading volume data, and technical indicator data; β0, β1,…, β n is the regression coefficient, which is determined by the training process of the linear regression algorithm; ε is the error term, which represents the part that the model cannot explain.
[0020] Preferably, the market price data is the historical price information of the target trading product, including the historical daily opening price, closing price, highest price and lowest price;
[0021] The transaction volume data is the historical transaction volume information of the target transaction product, including the historical daily buy transaction volume and sell transaction volume;
[0022] The technical indicator data are the moving average value, relative strength index value and MACD indicator value calculated based on market price data and transaction volume data.
[0023] Preferably, the functions of the data cleaning module include:
[0024] For abnormal values in historical price information, including the closing price of a certain day being higher than the highest price of the day or lower than the lowest price of the day, such data will be marked as abnormal and corrected or removed;
[0025] For historical trading volume information, if the sum of the buy volume and sell volume on a certain day does not match the total trading volume on that day, the corresponding trading volume data will be adjusted to ensure data consistency;
[0026] In the technical indicator data, if the moving average value, relative strength index value or MACD indicator value is extremely abnormal due to errors in the original data, including a large and instantaneous jump in the moving average value, such abnormal technical indicator values will be corrected or eliminated.
[0027] Preferably, the data preprocessing module is used to format and standardize the cleaned historical price information, historical transaction volume information and technical indicator data;
[0028] Formatting is used to convert data into a unified format, including unifying the date format and retaining the specified number of decimal places for price and volume data;
[0029] Standardization is used to scale the data so that it falls into a small specific range, including converting price data into relative price change values based on a certain point in time, so as to eliminate the impact of dimension and the influence of the variable's own variation in subsequent analysis.
[0030] Preferably, the strategy logic definition module defines the basic logic and rules of the trading strategy according to business needs and market characteristics, including defining the trading time window, selection criteria for trading products, trigger conditions for buying and selling, and setting principles for stop loss and take profit.
[0031] Preferably, the strategy verification module combines the trading signals generated by the trading prediction model to simulate the execution of the defined trading strategy, including buying, selling, stop loss and take profit operations according to the strategy rules; during the simulated trading process, the strategy verification module records the execution of each simulated transaction in detail, including the trading time, trading price, trading volume and trading results, and conducts a comprehensive statistical analysis of the results of the simulated transactions to evaluate the profitability of the trading strategy under the generated trading signals.
[0032] Preferably, the trading instruction generation module receives a trading signal from the trading strategy layer, wherein the trading signal is a buy or sell instruction obtained based on market data and trading strategy logic analysis; based on the received trading signal, the trading instruction generation module generates a specific trading instruction in combination with the real-time data of the current market, wherein the trading instruction includes the buy or sell quantity, price and transaction type; and finally sends the generated trading instruction to the trading instruction execution module;
[0033] The transaction instruction execution module is responsible for receiving the specific transaction instructions transmitted by the transaction instruction generation module and immediately connecting with the trading system; according to the type and content of the transaction instruction, the transaction instruction execution module automatically sends the corresponding buy and sell request to the exchange.
[0034] Preferably, the risk assessment module collects and analyzes historical performance data of trading strategies, identifies and quantifies the risk exposure of strategies under different market environments, including market risk, liquidity risk and operational risk; the risk response module provides a risk threshold setting function. During the transaction execution process, the module monitors market dynamics and transaction conditions in real time. Once the preset risk threshold is triggered, the corresponding risk response measures will be immediately initiated. The risk response measures include automatic closing of positions, reducing positions and suspending transactions to keep transaction risks within a controllable range.
[0035] Compared with the prior art, the present invention has the following beneficial effects:
[0036] By introducing a machine learning layer, the system is able to use complex machine learning algorithms to conduct in-depth analysis of historical market data, thereby more accurately predicting market dynamics and trading signals, improving the accuracy of trading decisions, and enabling traders to respond to market changes more quickly.
[0037] The trading strategy layer of this system can automatically generate trading signals based on the trained machine learning model and current market data, and then automatically execute buy and sell operations through the trading execution layer. This automated and intelligent trading method not only reduces the possibility of human intervention and misoperation, but also greatly improves trading efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is the overall structure diagram of the present invention. DETAILED DESCRIPTION
[0039] The following will be combined with the drawings in the embodiments 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.
[0040] See also Figure 1 The present invention provides a technical solution: a transaction execution control processing system based on trading strategies, whose overall architecture includes a data collection layer, a machine learning layer, a trading strategy layer, a trading execution layer, a risk control layer and a user interaction layer.
[0041] The data collection layer consists of a data acquisition module, a data cleaning module, and a data preprocessing module. The data acquisition module captures the market data of local currency transactions in real time from multiple data sources, such as the China Foreign Exchange Trade System (CFETS) financial information service platform and data published by exchanges. These data include but are not limited to market information such as price, trading volume, buying and selling direction, and counterparties. The data cleaning module cleans the captured raw market data, removes duplicate, erroneous or abnormal data, and ensures the accuracy and reliability of the data. The data preprocessing module further formats and standardizes the cleaned data to unify the data format, which is convenient for subsequent machine learning and data analysis.
[0042] The machine learning layer consists of a data set partitioning module, a model selection and construction module, a model training module, and a model evaluation and tuning module. The data set partitioning module scientifically divides the preprocessed historical data into training sets, validation sets, and test sets for model training, evaluation, and tuning. The model selection and construction module selects appropriate machine learning algorithms to build a transaction prediction model based on data characteristics and business needs. The model can predict transaction signals in real time based on current market data. The model training module uses the training set data to train the transaction prediction model and improves the model's prediction accuracy through continuous iteration and optimization. The model evaluation and tuning module uses the validation set and test set to evaluate the trained model and performs necessary tuning based on the evaluation results to ensure the performance of the model in actual applications.
[0043] The trading strategy layer includes a strategy logic definition module, a strategy signal generation module and a strategy verification module. The strategy logic definition module defines the core logic and rules of the trading strategy according to specific business needs and market characteristics. The strategy signal generation module generates trading signals based on the trained machine learning model and real-time market data. The strategy verification module uses historical data and trading signals generated by the trading prediction model to backtest and verify the defined trading strategy. Verified favorable trading signals will be passed to the trading execution layer.
[0044] The transaction execution layer consists of a transaction instruction generation module, a transaction instruction execution module and a transaction record management module. The transaction instruction generation module generates specific transaction instructions, including buying and selling directions, quantities, prices, etc., based on the favorable transaction signals output by the transaction strategy layer. The transaction instruction execution module is responsible for connecting with the exchange's interface and executing specific buying and selling operations. The transaction record management module is responsible for recording detailed information on each transaction, including transaction time, price, quantity, results, etc., for subsequent analysis and inquiry.
[0045] The risk control layer includes a risk indicator calculation module, a risk assessment module and a risk response measures module. The risk indicator calculation module calculates and monitors various risk indicators in real time. The risk assessment module evaluates and warns of potential risks based on real-time market data and transaction conditions. The risk response measures module formulates corresponding risk response measures based on the results of risk assessment to reduce or avoid potential risks.
[0046] The user interaction layer consists of a real-time market information display module, a trading strategy adjustment module, a trading status monitoring module and a historical trading record query module. The real-time market information display module will provide users with the latest market information to help users better grasp market dynamics. The trading strategy adjustment module allows users to adjust the parameters of trading strategies according to actual conditions to meet personalized trading needs. The trading status monitoring module will display the status and results of transactions in real time, allowing users to keep abreast of the latest trading situations at any time. The historical trading record query module provides a historical trading record query function to facilitate users to conduct trading analysis and summary.
[0047] The present invention will be further described below in conjunction with Examples 1 to 3:
[0048] Embodiment 1:
[0049] The present invention constructs a transaction prediction model based on a linear regression algorithm, and through precise data analysis and model training, it is expected to improve the overall performance of transaction execution. According to the characteristics of historical transaction data, this algorithm can effectively describe the relationship between independent variables such as market prices, transaction volumes, technical indicators, and the dependent variable of transaction signals (including buying and selling directions, order prices and quantities). With the help of the constructed prediction model, the system will calculate the transaction signals in real time and quickly execute transaction operations based on these signals, thereby ensuring the efficiency and accuracy of the transaction.
[0050] 10. The linear regression algorithm uses the following formula to describe the relationship between the independent variable and the dependent variable:
[0051] Y=β0+β1X1+β2X2+…+β n X n +ε
[0052] Among them, Y is the dependent variable, representing the predicted trading signal, which includes the buying and selling direction, the price and quantity of the order; X1, X2, …, X n is the independent variable, representing the input market data, including market price data, trading volume data, and technical indicator data; β0, β1,…, β n is the regression coefficient, which is determined by the training process of the linear regression algorithm; ε is the error term, which represents the part that the model cannot explain.
[0053] The market price data is the historical price information of the target trading product, including the historical daily opening price, closing price, highest price and lowest price; the opening price refers to the price of the target trading product at the beginning of each trading day. The opening price is usually determined by the call auction before the market opens, reflecting the preliminary pricing of the trading product by the market on that day; the closing price refers to the price of the target trading product at the end of each trading day. The closing price is the final confirmation of the value of the trading product by the market on that day, which is of great significance for analyzing market trends and predicting future price trends. The highest price refers to the highest price reached by the target trading product in each trading day. The highest price reflects the highest degree of market recognition of the trading product on that day and is an important indicator for analyzing market fluctuations and trader sentiment. The lowest price refers to the lowest price reached by the target trading product in each trading day. The lowest price shows the lowest degree of market acceptance of the trading product on that day, which helps to identify the market's support level and potential risk points.
[0054] The transaction volume data is the historical transaction volume information of the target transaction product, including the historical daily buy volume and sell volume; buy volume refers to the total volume of transactions completed in the buy direction on each trading day. Buy volume reflects the strength of the market's buyer power and has reference value for judging market trends and predicting future price changes. Sell volume refers to the total volume of transactions completed in the sell direction on each trading day. Sell volume shows the activeness of the market's seller power and helps analyze the supply and demand relationship in the market and traders' willingness to sell.
[0055] The technical indicator data are the moving average value, relative strength index value and MACD indicator value calculated based on the market price data and trading volume data; moving average value: by calculating the average price of the target trading product over a period of time and drawing a line chart to show the long-term trend of the market price, the moving average helps to smooth price fluctuations and highlight the main trend of the market; relative strength index value: by comparing the average closing price changes of the rising days and falling days of the target trading product, the overbought or oversold state of the market is evaluated. The relative strength index helps to identify the turning points and potential reversal signals of the market; MACD indicator value: by calculating the difference between the short-term and long-term exponential moving averages of the target trading product, the momentum and trend changes of the market are revealed. The MACD indicator helps to capture short-term market fluctuations and trading opportunities.
[0056] The data set partitioning module is used to partition the preprocessed historical data into a training set, a validation set, and a test set; the training set occupies most of the data and is used for model training; the validation set is used for model selection and parameter adjustment; and the test set is used to evaluate the final performance of the model, with a partition ratio of 70%:15%:15%.
[0057] The model training module uses the training set data to train the transaction prediction model; before the training begins, the transaction prediction model needs to be initialized, including setting the initial parameters of the model (such as regression coefficients, intercepts, etc.), and selecting appropriate loss functions and optimization algorithms.
[0058] After initialization, the training set data is input into the initialized trading prediction model. These data are used as samples for model training and are used to adjust the parameters of the model. According to the current parameters of the model, the output of the input data after passing through the model is calculated, and the selected loss function (such as the mean square error loss function) is used to calculate the difference between the model output and the actual trading signal. According to the calculated loss, the optimization algorithm (such as the gradient descent method) is used to adjust the parameters of the model. Repeat the above calculation steps until the preset number of training rounds (epochs) is reached or the performance of the model reaches a certain predetermined standard. During the model training process, the appropriate learning rate and number of iterations should be selected to avoid overfitting or underfitting the model to the training data. Regularization techniques (such as L1 regularization and L2 regularization) can be used to enhance the generalization ability of the model and prevent the occurrence of overfitting.
[0059] The model evaluation and tuning module uses the validation set and the test set to evaluate the performance of the transaction prediction model;
[0060] Model evaluation: First, input the validation set data into the model to obtain the model's prediction results; calculate various evaluation indicators, including accuracy, recall, and F1 score, based on the prediction results and actual transaction results; analyze the evaluation indicators to preliminarily judge the performance of the model. In the same way, use the test set data to further evaluate the model. The purpose of the test set evaluation is to verify the model's generalization ability on unknown data.
[0061] Model tuning: Based on the evaluation results of the validation set and test set, deeply analyze the performance bottlenecks and potential problems of the model. Adjust the model's hyperparameters, including learning rate, regularization coefficient, and batch size; use cross-validation techniques (such as k-fold cross-validation) to more accurately evaluate model performance; retrain the model based on the tuned parameters and configuration; use new validation sets and test sets to evaluate the tuned model. Compare and analyze the model performance before and after tuning to evaluate the tuning effect.
[0062] Embodiment 2:
[0063] The data preprocessing module is responsible for formatting and standardizing the cleaned historical price information, historical trading volume information, and technical indicator data to provide a unified and standardized data set for subsequent data analysis and model training. The following is a detailed and expanded description of the implementation of the data preprocessing module:
[0064] Formatting mainly converts data into a unified format to ensure the standardization and consistency of data:
[0065] Standardize date formats: Convert date data from different sources or in different formats to a unified date format, such as "YYYY-MM-DD" or "YYYYMMDD". For data in timestamp format, you can convert it to the corresponding date format.
[0066] Price and volume data formatting: Price and volume data are usually formatted to a specified number of decimal places, such as two decimal places for prices and integer places for volumes. Data is formatted by rounding or truncating.
[0067] Standardization aims to scale the data so that it falls between 0 and 1. This can eliminate the impact of the dimension and the size of the variable's own variation, making different features comparable:
[0068] Price data standardization:
[0069] For price data, the following formula can be used for standardization:
[0070]
[0071] Among them, x is the original price data, μ is the mean of the price data, and σ is the standard deviation of the price data.
[0072] Volume data normalization:
[0073] Considering that the fluctuation range of trading volume may be large, the Min-Max normalization method is used to scale the trading volume data to the interval [0,1]. The formula is as follows:
[0074]
[0075] Among them, x is the original trading volume data, min and max are the minimum and maximum values in the trading volume data respectively.
[0076] Embodiment 3:
[0077] The main function of the strategy logic definition module is to define the basic logic and rules of the trading strategy based on business needs and market characteristics.
[0078] Define the time window for trading: The strategy logic definition module first determines the time window for the trading strategy, including the start and end time of the trading day, and possible trading frequencies (such as intraday trading, overnight trading, etc.). The time window can be set based on the market's opening and closing times, or it can take into account the market's active hours and liquidity conditions.
[0079] Selection criteria for trading products: Based on business needs and market analysis, the strategy logic definition module will set the selection criteria for trading products, including factors such as volatility, liquidity, and market influence of the product. For example, you can choose a contract with moderate volatility and good liquidity as the trading object.
[0080] Trigger conditions for buying and selling: The strategy logic definition module will clearly define the trigger conditions for buying and selling.
[0081] Principles for setting stop loss and take profit: In order to control risks and lock in profits, the strategy logic definition module will set the principles of stop loss and take profit. The stop loss point is usually set at a certain percentage or fixed price below the purchase price to limit possible losses, and the take profit point is set at the price after the expected profit is achieved to ensure timely exit from the market before the price pulls back.
[0082] The strategy signal generation module generates trading signals based on the trained machine learning model and real-time market data. The module obtains market prices, trading volumes and other relevant data in real time through the data interface, performs pre-processing operations such as cleaning and normalization on the real-time data to meet the input requirements of the machine learning model, and inputs the pre-processed real-time data into the trained machine learning model to generate buy, sell or hold trading signals based on the output of the model.
[0083] The strategy verification module combines the trading signals generated by the trading prediction model to simulate the execution of the defined trading strategy, including buying, selling, stop loss and take profit operations according to the strategy rules; during the simulated trading process, the strategy verification module records the execution of each simulated transaction in detail, including the trading time, trading price, trading volume and trading results, and conducts a comprehensive statistical analysis of the results of the simulated transactions to evaluate the profitability of the trading strategy under the generated trading signals. The trading signals verified as favorable by the strategy verification module will be passed to the trading execution layer to ensure that only verified valid signals will be actually executed, thereby improving the robustness and profitability of the transaction.
[0084] The trading instruction generation module receives trading signals from the trading strategy layer, and the trading signals are buy and sell instructions obtained based on market data and trading strategy logic analysis; based on the received trading signals, the trading instruction generation module generates specific trading instructions in combination with the real-time data of the current market, and the trading instructions include the quantity, price and transaction type of buying or selling; and finally sends the generated trading instructions to the trading instruction execution module;
[0085] The transaction instruction execution module is responsible for receiving the specific transaction instructions transmitted by the transaction instruction generation module and immediately connecting with the trading system; according to the type and content of the transaction instruction, the transaction instruction execution module automatically sends the corresponding buy and sell request to the exchange.
[0086] The risk assessment module collects and analyzes the historical performance data of trading strategies, identifies and quantifies the risk exposure of strategies under different market environments, including market risk, liquidity risk and operational risk; the risk response module provides a risk threshold setting function. During the transaction execution process, the module monitors market dynamics and transaction conditions in real time. Once the preset risk threshold is triggered, the corresponding risk response measures will be immediately initiated. The risk response measures include automatic closing of positions, reducing positions and suspending transactions to keep transaction risks within a controllable range.
[0087] The risk indicator calculation module calculates and monitors various risk indicators; the risk assessment module assesses potential risks based on market data and transaction conditions; the risk response module formulates corresponding measures based on the risk assessment results;
[0088] The real-time market information display module provides the latest market information to users; the trading strategy adjustment module allows users to adjust trading strategy parameters according to actual conditions; the trading status monitoring module displays the trading status in real time; and the historical trading record query module provides a historical trading record query function.
[0089] The implementation of the real-time market information display module mainly depends on the following steps:
[0090] (1) Data acquisition: Real-time acquisition of the latest market data from the data collection layer, including key information such as price, transaction volume, buy and sell orders, etc.
[0091] (2) Data processing: Perform necessary processing on the acquired raw data, such as formatting, screening and sorting, to ensure the accuracy and readability of the data.
[0092] (3) Data display: The processed data is displayed to users in an intuitive way through graphical interfaces such as candlestick charts, bar charts, pie charts, etc. At the same time, real-time updated data streams are provided to ensure that users can keep abreast of market trends at any time.
[0093] The implementation steps of the trading strategy adjustment module are as follows:
[0094] (1) Strategy parameter definition: Based on the requirements of the trading strategy, a series of adjustable parameters are predefined, such as stop loss point, take profit point, trading frequency, etc.
[0095] (2) Parameter adjustment interface: Design a user-friendly interface that allows users to adjust the above parameters according to actual conditions. The interface can be in the form of a slider, input box, etc. to facilitate users to make fine adjustments.
[0096] (3) Parameter transfer and update: After the user adjusts the parameters, the module passes the new parameter values to the trading strategy layer to ensure that the trading strategy can be updated in real time based on the user's adjustments.
[0097] The implementation of the transaction status monitoring module is as follows:
[0098] (1) Status information acquisition: The current transaction status information is obtained from the transaction execution layer in real time, including status such as completed, uncompleted, and cancelled orders.
[0099] (2) Status display: The current transaction status is displayed to the user in real time through a graphical interface or text form. For abnormal status, such as transaction failure, timeout, etc., a striking method can be used to prompt.
[0100] The implementation steps of the historical transaction record query module are as follows:
[0101] (1) Data storage: The transaction record management module stores detailed information of each transaction in the database, including key information such as transaction time, transaction type, transaction quantity, and transaction price.
[0102] (2) Query interface design: Design a user-friendly query interface that allows users to filter and query based on time range, trading type, and other conditions.
[0103] (3) Result display: According to the user's query conditions, the corresponding transaction records are retrieved from the database and displayed to the user in the form of tables or charts. At the same time, an export function is provided to facilitate users to further analyze and process the data.
[0104] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0105] 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. A transaction execution control processing system based on a transaction strategy, characterized in that: It includes data collection layer, machine learning layer, trading strategy layer, trading execution layer, risk control layer and user interaction layer; The data collection layer includes a data acquisition module, a data cleaning module and a data preprocessing module; The data acquisition module is used to capture the market data of local currency transactions from multiple data sources in real time; the data cleaning module is responsible for cleaning the captured data; the data preprocessing module is used to format and standardize the data; The machine learning layer includes a data set partitioning module, a model selection and construction module, a model training module, and a model evaluation and tuning module; The data set division module is used to divide the pre-processed historical data into a training set, a validation set and a test set; the model selection and construction module uses a machine learning algorithm to build a transaction prediction model, and the transaction prediction model is used to predict transaction signals in real time based on current market data; the model training module uses the training set data to train the transaction prediction model; the model evaluation and tuning module uses the validation set and the test set to evaluate the performance of the transaction prediction model; The trading strategy layer includes a strategy logic definition module, a strategy signal generation module and a strategy verification module; The strategy logic definition module defines the core logic and rules of the trading strategy according to business needs; the strategy signal generation module generates trading signals based on the trained machine learning model and current market data; the strategy verification module uses historical data and trading signals generated by the trading prediction model to backtest and verify the defined trading strategy, and transmits the verified favorable trading signals to the trading execution layer; The transaction execution layer includes a transaction instruction generation module, a transaction instruction execution module and a transaction record management module; The trading instruction generation module generates specific trading instructions according to the trading signals output by the trading strategy layer; the trading instruction execution module is responsible for connecting with the exchange interface to execute buying and selling operations; the trading record management module records the detailed information of each transaction; The risk control layer includes a risk indicator calculation module, a risk assessment module and a risk response measure module; The risk indicator calculation module calculates and monitors various risk indicators; the risk assessment module assesses potential risks based on market data and transaction conditions; the risk response module formulates corresponding measures based on the risk assessment results; The user interaction layer includes a real-time market information display module, a trading strategy adjustment module, a trading status monitoring module and a historical trading record query module; The real-time market information display module provides the latest market information to users; the trading strategy adjustment module allows users to adjust trading strategy parameters according to actual conditions; the trading status monitoring module displays the trading status in real time; and the historical trading record query module provides a historical trading record query function.
2. The transaction execution control processing system based on transaction strategy according to claim 1, characterized in that: The model selection and construction module uses a linear regression algorithm to construct a transaction prediction model. The linear regression algorithm uses the following formula to describe the relationship between the independent variable and the dependent variable: Y=β0+β1X1+β2X2+…+β n X n +e Among them, Y is the dependent variable, representing the predicted trading signal, which includes the buying and selling direction, the price and quantity of the order; X1, X2, …, X n is the independent variable, representing the input market data, including market price data, trading volume data, and technical indicator data; β0, β1,…, β n is the regression coefficient, which is determined by the training process of the linear regression algorithm; ε is the error term, which represents the part that the model cannot explain.
3. The transaction execution control processing system based on transaction strategy according to claim 2, characterized in that: The market price data is the historical price information of the target trading product, including the historical daily opening price, closing price, highest price and lowest price; The transaction volume data is the historical transaction volume information of the target transaction product, including the historical daily buy transaction volume and sell transaction volume; The technical indicator data are the moving average value, relative strength index value and MACD indicator value calculated based on market price data and transaction volume data.
4. The transaction execution control processing system based on transaction strategy according to claim 3, characterized in that: The functions of the data cleaning module include: For abnormal values in historical price information, including the closing price of a certain day being higher than the highest price of the day or lower than the lowest price of the day, such data will be marked as abnormal and corrected or removed; For historical trading volume information, if the sum of the buy volume and sell volume on a certain day does not match the total trading volume on that day, the corresponding trading volume data will be adjusted to ensure data consistency; In the technical indicator data, if the moving average value, relative strength index value or MACD indicator value is extremely abnormal due to errors in the original data, including a large and instantaneous jump in the moving average value, such abnormal technical indicator values will be corrected or eliminated.
5. The transaction execution control processing system based on transaction strategy according to claim 4, characterized in that: The data preprocessing module is used to format and standardize the cleaned historical price information, historical transaction volume information and technical indicator data; Formatting is used to convert data into a unified format, including unifying the date format and retaining the specified number of decimal places for price and volume data; Standardization is used to scale the data so that it falls into a small specific range, including converting price data into relative price change values based on a certain point in time, so as to eliminate the impact of dimension and the influence of the variable's own variation in subsequent analysis.
6. The transaction execution control processing system based on transaction strategy according to claim 5, characterized in that: The strategy logic definition module defines the basic logic and rules of the trading strategy according to business needs and market characteristics, including defining the trading time window, selection criteria for trading products, trigger conditions for buying and selling, and setting principles for stop loss and take profit.
7. The transaction execution control processing system based on transaction strategy according to claim 6, characterized in that: The strategy verification module combines the trading signals generated by the trading prediction model to simulate the execution of the defined trading strategy, including buying, selling, stop loss and take profit operations according to the strategy rules; during the simulated trading process, the strategy verification module records the execution of each simulated transaction in detail, including the trading time, trading price, trading volume and trading results, and conducts a comprehensive statistical analysis of the results of the simulated transactions to evaluate the profitability of the trading strategy under the generated trading signals.
8. The transaction execution control processing system based on transaction strategy according to claim 7, characterized in that: The trading instruction generation module receives trading signals from the trading strategy layer, and the trading signals are buy and sell instructions obtained based on market data and trading strategy logic analysis; based on the received trading signals, the trading instruction generation module generates specific trading instructions in combination with the real-time data of the current market, and the trading instructions include the quantity, price and transaction type of buying or selling; and finally sends the generated trading instructions to the trading instruction execution module; The transaction instruction execution module is responsible for receiving the specific transaction instructions transmitted by the transaction instruction generation module and immediately connecting with the trading system; according to the type and content of the transaction instruction, the transaction instruction execution module automatically sends the corresponding buy and sell request to the exchange.
9. The transaction execution control processing system based on transaction strategy according to claim 8, characterized in that: The risk assessment module collects and analyzes the historical performance data of trading strategies, identifies and quantifies the risk exposure of strategies under different market environments, including market risk, liquidity risk and operational risk; the risk response module provides a risk threshold setting function. During the transaction execution process, the module monitors market dynamics and transaction conditions in real time. Once the preset risk threshold is triggered, the corresponding risk response measures will be immediately initiated. The risk response measures include automatic closing of positions, reducing positions and suspending transactions to keep transaction risks within a controllable range.
Citation Information
Patent Citations
Multi-module automatic trading system based on network distributed computing
CN106934716A
Stock quantitative transaction method based on deep reinforcement learning, storage medium and equipment
CN111861751A
Stock transaction decision-making method based on Kailey criterion and loss stopping strategy
CN115660858A
Financial transaction method and device based on reinforcement learning, equipment and storage medium
CN118279050A
Cited By
Transaction sharing method based on delivery service platform
CN119558976A
Financial transaction analogue simulation method and system based on VR
CN119718090A
A trading simulation backtesting framework and system
CN119739607A
Multi-factor quantitative transaction modeling and automatic execution platform based on artificial intelligence
CN121329480A
