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

针对当前方法对金融市场长、短期趋势的预测结果偏差较大的问题,提出了基于机器学习的金融市场趋势预测方法。依照一定时间间隔采集金融数据,产生金融数据时间序列,采用小波分析方法对金融数据时间序列进行处理,去除其中的噪声,保留近似数据。通过卷积长短期记忆神经网络对金融数据时间序列进行学习,建立金融市场预测模型。测试结果表明,所提出的方法可以有效清除噪声、平滑初始数据,在短期趋势与长期趋势预测中均具有较高预测精度。

In view of the large deviation of the current methods in the prediction results of the long-term and short-term trends of the financial market, a financial market trend prediction method based on machine learning algorithm is proposed. The financial data are collected according to a certain time interval to generate the financial data time series. The wavelet analysis method is used to process the financial data time series to remove the noise and retain the approximate data. The time series of financial data is studied by convolution long-term and short-term memory neural network, and the financial market prediction model is established. The test results show that this method can effectively remove noise and smooth the initial data, and has high prediction accuracy in short-term trend and long-term trend prediction.