Application of data mining technology on particle swarm optimization and support vector regression in shareprice prediction

Precise forecasting for shareprice is very important to investment and financing. Support vector regression, called as SVR, is a novel learning algorithm based on statistical learning theory, which has greater generalization ability than traditional neural networks. In order to select the appropriate parameters of SVR, particle swarm optimization is introduced to choose the user-determined parameters of SVR here. Therefore, data mining technology on particle swarm optimization and support vector regression is presented to shareprice prediction. Closingprice of 23 trading days of routon electronic is applied to testify the feasibility of the proposed method in the shareprice forecasting. The experiment results demonstrate that the proposed algorithm is better than the traditional shareprice forecasting algorithm.