GDP Forecasting Based on Online Weighted Least Squares Support Vector Machine

In order to improve the prediction accuracy of chaotic time series, a chaotic time series forecasting method based on online weighted least squares support vector machine regression (WLS-SVM) is proposed. In this method, a sliding time window is built and data in the sliding time window are employed to construct the dynamic model of a system. The model of the system is refreshed on-line with the rolling of the time window. In order to make full use of data information, different weights are assigned to different data in the sliding time window. Online WLS-SVM prediction method is applied to predict the GDP data in macro economy. Results show that the proposed method can be realized easily and has good performance in robustness and precision in chaotic time series prediction.