基于粗糙集和支持向量机的股指期货预测模型研究

本文提出基于粗糙集和支持向量机的股指期货走势预测模型。在模型中首先使用粗糙集对指标集进行特征选择,剔除冗余指标,然后使用支持向量机对基于历史数据的股指期货价格走势进行预测。为了评估该预测模型的性能,将预测结果与传统的自回归移动平均模型和BP神经网络模型的预测结果进行比较。实验结果表明了该模型的有效性。

Research on a Rough Sets and Support Vector Machines Based Stock Index Futures Prediction Model

This paper presents a rough set （RS） and support vector machines （SVM） based stock index futures prediction model. This model employs a rough set to select the eigenvectors and to remove some redundancy eigenvectors, and then employs SVM to predict the historical data based stock index futures price tendency. We compare its performance with that of conventional ARMA model and neural network model to evaluate its prediction capability. Experimental results show that this model has better prediction accuracy, so it is a promising stock index futures prediction tool.