基于概率后缀树的股票时间序列预测方法研究

Research of Stock Time Series Based on probabilistic Suffix Tree

摘要:

本文在时间序列符号化基础上，引入概率后缀树（Probabilistic Suffix Tree，PST）模型，构建基于时间序列符号化和概率后缀树相结合的股票预测模型。本文选择在沪深300的10股票数据上将预测模型与传统的马尔科夫模型(Markov Model，MM)和自回归移动平均模型(Auto Regressive Moving Average Model，ARMA)进行对比，结果显示本文提出的股票预测模型优于MM模型和ARMA模型，验证了本文所提出的预测模型在投资收益上的有效性。

Abstract:

this paper introduces a Probabilistic Suffix Tree (PST) method based on the time series symbolization, and constructs a stock forecasting model based on the combination of time series symbolizationa and PST. In addition, the Markov Model MM and the Auto Regressive Moving Average Model (ARMA) are compared with the forecasting model of this paper.The stock of 10 CSI 300 indices is used as the experimental sample. The results show that the stock forecasting model proposed in this paper is better than the MM model and the ARMA model，and proves the validity of the forecasting model proposed in this paper.