초록 열기/닫기 버튼
Support vector quantile regression (SVQR) is capable of providing more complete description of the linear and nonlinear relationships among random variables. In this paper we propose the sparse SVQR whose objective function is composed of a weighted quadratic loss function and l1 norm penalty term. We use the iterative reweighted least squares (IRWLS) procedure to solve the objective problem of the proposed SVQR. Furthermore, we introduce the generalized approximate cross validation function to select the hyper-parameters which affect the performance of SVQR. Experimental results are then presented, which illustrate the performance of the sparse SVQR using IRWLS procedure.
