Population Forecasting Model Based on Wavelet Least Squares Support Vectors Machines

Abstract

A new population forecasting model based on wavelet least squares support vector machines (WLS-SVM) was proposed in this paper. Support vector machines (SVM) is a new kind of neural networks based on statistical learning theory, which is a powerful tool to deal with problems such as small sample, nonlinearity, high dimension, and local minima. LS-SVM is a refined SVM which use equality instead of inequality constraints of standard SVM to simplify calculation of parameters of SVM. Wavelet kernel function was used to construct WLS-SVM in this paper. Because this kind of kernel function can simulate almost any function in square integral space, it enhances the generalization ability of the SVM. Based on wavelet kernel function, WLS-SVM regression model was proposed. Then the WLS-SVM regression model deduced a new population forecasting method. Experiments on predication of population of China show the proposed method has higher precision than classical neural networks such as radial basis function (RBF) neural networks.