一种基于时间序列分析的股票走势预测模型  

A Novel Stock Trend Prediction Model Based on Time Series Analysis

作　　者：李奋华[1,2] 赵润林[1] 

机构地区：[1]运城学院计算机科学与技术系,山西运城044000 [2]中国科学院大学,北京100190

出　　处：《现代计算机（中旬刊）》2016年第7期14-17,共4页Modern Computer

基　　金：国家自然科学基金项目(No.61272480)

摘　　要：在信息爆炸时代,在股市中积累的具有时间标签的股票交易数据越来越多,仅仅依靠传统手工的股票数据分析办法无法有效地获取对投资者有价值的知识。为了能够从海量股票历史数据中更好地获取对投资者有用的信息,高效地指导投资者投资,同时,为股票市场管理提供有效的决策支持,在股票分析中引入数据挖掘技术,提出一种基于时间序列的股票走势预测模型,在真实股票数据集上的实验表明,该模型对股票走势的预测具有较好的效果。In information explosion era, there are the massive stock exchange data being stored in the computer systems in stock market. However,the valuable knowledge for the investors is not obtained if there are only some traditional and manual analysis methods for stock data. It is very vital to find the potential and useful information for many investors from the massive stock data, which can instruct the investors and the stock market management decision effectively. Applies data mining technologies to stock analysis, proposes a novel stock trend pre-diction model based on time series analysis. Through the experiments on real stock exchange datasets, some empirical studies are shown to demonstrate the effectiveness of this model on the real stock exchange datasets.

关 键 词：信息爆炸 数据挖掘 时间序列分析 股票预测 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]