Wavelet support vector machines and multi-elitist particle swarm optimization for time series forecasting

Abstract

Listen

Translate article

In this paper, we present a new method for time series forecasting based on wavelet support vector machines (WSVM). To better represent any curve in L^2(R^n) space (quadratic continuous integral space), we used a new kernel function. This function is the wavelet function. The SVM with wavelet kernel function is referred to as a wavelet SVM. In order to determine the optimal parameter of the WSVM, the multi-elitist particle swarm optimization (PSO) was used. Computational results demonstrate the effectiveness of the proposed method over the traditional methods.