Twin Support Vector Regression Model Based on Heteroscedastic Gaussian Noise and Its Application

The main purpose of twin support vector regression (TSVR) is to find linear or nonlinear relationships in sample data, and then predict future data. TSVR the decomposition a large convex quadratic programming problem into two small problems. Therefore, not only has advantages fast computation low computational complexity, but also better performance. Classic SVR, assuming that accords with mean zero, variance noise gaussian distribution consider effects noise, some practical applications, such as wind speed forecasting, characteristic more line for heteroscedasticity distribution, therefore, return existing technology best. In this paper, characteristics Gaussian are introduced model TSVR, based on (TSVR-HGN) constructed. Lagrange multiplier method used solve problem, optimization algorithm global optimization. artificial data set, UCI set were selected experimental comparison. results show TSVR-HGN prediction accuracy.