基于粗集支持向量机的区域经济走势预测方法

Forecasting Regional Economic Tendency Using Rough Sets-Based Support Vector Machines

针对区域经济预测中多属性而支持向量机方法无法有效选择的问题,提出了一个基于粗集理论和支持向量机的区域经济走势预测方法。方法利用粗集理论在处理多属性数据方面的优势对区域经济走势预测的条件属性进行约简,约简后的数据进入支持向量机的预测系统,从而减少了支持向量机的训练数据,在一定程度上克服了支持向量机方法处理速度慢的缺点。将方法应用于某区域经济走势预测中,获得了较好的预测结果。实证结果表明,方法具有较好的预测能力,与标准支持向量机方法相比,方法具有明显的优势。

In order to resolve the problem of how to select the variables for support vector machines （SVM） of regional economic forecasting, a hybrid forecasting method, rough sets based support vector machines （RSSVM） model, is presented for forecasting regional economic tendency. In this hybrid approach, rough sets （RS） are used for variable selection in order to reduce the model complexity of support vector machines （ SVM ） and improve the speed of SVM, and then the SVM is used to identify regional economic movement direction based on the historical data. The empirical results reveal that RSSVM method has understanding forecasting ability. Compared with the standard SVM, RSSVM has great superiority in predicting accuracy.