循环神经网络股票预测

Recurrent Neural Network for Stock Forecast

现有文献中主要讨论了传统机器学习方法在股票预测上的应用,对于循环神经网络尝试较少且均未在日级以下数据上尝试预测。本文采用标普500每分钟数据来预测股票价格,利用循环神经网络做模型构建,分别对未来一分钟和十分钟不同时间区间做出预测,从而说明分钟级股价的规律性。

Existing literature mainly discusses the application of traditional machine learning method in stock prediction.And sel-dom attempt to use recurrent neural network and never using the minutes data.To illustrate the regularity of the stock price at theminute level.This article uses the data of the S P 500 per minute to predict the stock price,using recurrent neural network as amodel to predict the future one and ten minutes in different time intervals respectively.