Stock Market Prediction Model Based on Genetic Algorithm and Support Vector Regression

摘要：

Accurate prediction of stock market is very important to obtain high profits from stock market As the operational parameters of support vector regression are usually determined by user, which leads to bad prediction results. In the paper, genetic algorithm and support vector regression (GA-SVR) model is developed to predict stock market, and genetic algorithm selects the best parameters of support vector regression. The comparison of the prediction error between GA-SVR and BP neural network to show the excellence of GA-SVR compared with BP neural network in the experimental analysis. It can be seen from the comparison results of the prediction error between GA-SVR and BP neural network that GA-SVR can gain the lower prediction error than BP neural network in the stock market prediction.

关键词： support vector regression prediction technology stock market efficient management neural network

作者: Bao Yilan

作者单位: Software Methodology and Software Engineering Dalian Maritime University Dalian 116026,China

会议类型: 国际会议

会议名称: 2010 3rd International Conference on Environmental and Computer Science(2010年第三届环境与计算机科学国际会议 ICECS 2010)

会议地点: 昆明

会议语种:英文

页码: 148-151

在线出版日期: 2010-10-17（万方平台首次上网日期，不代表论文的发表时间）