The Optimization of Share Price Prediction Model Based on Support Vector Machine

In recent years, the prediction for variation trend of share price is a hot issue which has drawn people´ attention. For share price is a group of non-linear time series data, the prediction accuracy of traditional prediction method is not high enough. The paper tries to bring the technology of support vector machine to the prediction model of share price to forecast the closing price on the third day. Besides, it optimizes the selection for each kind of parameter in the model by particle swarm optimization (PSO). The experiment result shows that the model of share price based on support vector machine which is optimized by particle swarm can predict the closing price of the stock on the third day precisely. This method has high actual value.