基于机器学习的中证500指数期货价格预测

Research on the Futures Price of China Securities 500 Index Based on Machine Learning

摘要: 文章采用多种时间序列模型对中证500指数期货收盘价数据展开分析，通过递增窗口交叉验证以及网格调参的方法，系统性地选择最优参数，以提高对期货收盘价数据的预测精度。研究中使用了机器学习中的随机森林、支持向量机、多层神经网络以及ARIMA模型作为基准模型，通过对比分析这些模型的预测效果，从而深入了解它们在对中证500指数期货收盘价时间序列上的性能表现。结果表明：对于该期货价格收盘价的性质和特质，随机森林进行时间序列预测比其他模型的预测精度更高。

Abstract: In this paper, multiple time series models are used to analyze the closing price data of CSI 500 index futures, and the optimal parameters are selected by increasing window cross-validation and network parameters to predict the closing price data. Random forest, support vector machine, multi-layer neural network and ARIMA model in machine learning are used as the benchmark model to compare and analyze the prediction effect. The results show that the time series prediction of random forest is more accurate than other models for the nature and characteristics of the closing price of the futures.