基于GA-Transformer模型的多因子股票预测

Multi-factor stock forecasting method based on a GA-Transformer model

股票市场中拥有大量用于描述股票价格变化的财务指标,这些指标为股票价格预测提供了良好的数据基础.但由于股票数据存在高维相关性和时序性等特点,导致精确预测股票价格存在困难.为提高股票价格预测精度,文章提出了基于GA-Transformer模型的多因子股票预测方法,该方法使用遗传算法(GA)进行特征选择,并结合Transformer模型进行股票预测,提升了模型特征抽取能力.实验结果表明,GA-Transformer模型在包括建设银行和贵州茅台等六支股票数据集上的预测表现均优于股票预测主流模型.

In the domestic stock market,there are a large number of financial indicators used to describe the change of stock prices,which provide a good data basis for stock price prediction.However,it is difficult to accurately predict stock prices because of a high dimensional correlation and the time sequence of stock data.In order to improve the accuracy of stock price prediction,this paper proposes a multi-factor stock prediction method based on a GA-Transformer model,which uses a genetic algorithm(GA)for feature selection and a Transformer model for stock prediction,and improves the ability of feature extraction.The experimental results show that the performance of the GA-Transformer model on six stock data sets,including CCB and Kweichow Moutai,is better than the mainstream models.