基于机器学习算法的金融市场趋势预测研究  

Research on financial market trend prediction based on machine learning algorithm

作　　者：刘博[1] LIU Bo(Chinese Academy of Social Sciences,Beijing 102488,China)

机构地区：[1]中国社会科学院,北京102488

出　　处：《现代电子技术》2022年第9期83-87,共5页Modern Electronics Technique

摘　　要：金融市场受到多种因素影响,具有强烈的非线性和时变性,当前方法无法准确描述金融市场趋势的变化特点,导致金融市场趋势预测偏差比较大,预测结果精度较低。为了提高金融市场趋势预测精度,设计了基于机器学习算法的金融市场趋势预测方法。采集金融市场趋势变化的历史样本数据,并对样本数据进行一定预处理;采用机器学习算法中的深度神经网络模型对历史数据进行建模;结合人工免疫算法和粒子群优化算法对深度神经网络预测模型的参数进行优化,构建金融市场趋势预测模型,输出最优解,完成金融市场趋势预测。测试结果表明,所提方法T值低于0.13、HR值高于0.95,对金融市场趋势预测精度更高,实用性更强。The financial market is affected by many factors,so it has strong nonlinearity and time variability.The current methods fail to accurately describe the changing characteristics of the financial market trend,which results in large deviation and low accracy of the prediction results in the financial market trend prediction.Therefore,a financial market trend prediction method based on machine learning algorithm is designed to improve the prediction accuracy.The historical sample data of financial market trend changes are collected and preprocessed.The deep neural network(DNN)model in machine learning algorithm is used to model the historical data.The parameters of DNN prediction model are optimized in combination with the artificial immune algorithm(AIA)and the particle swarm optimization(PSO)algorithm,so as to construct the financial market trend prediction model and output the optimal solution.On the basis of the above,the trend prediction of financial market is completed.The test results show that the T value of the proposed method is lower than 0.13 and its HR value is higher than 0.95,so it can be seen that the proposed method has higher accuracy and practicability in predicting the trend of financial market.

关 键 词：趋势预测 金融市场 机器学习算法 参数优化 动态预测 预测模型 

分 类 号：TN911.1-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]