The estimates in Partially Linear Models have been studied previously in tradi- tional smoothing methods such as smoothing spline, kernel and piecewise polynomial smoothers. Here, we apply the regularized wavelet estimators by penalizing the L1 norm of the wavelet coe±cients of the nonparametric function. The regularization parameter is chosen by universal threshold. Simulation results show that regularized wavelet ap- proach performs well. The wavelet method makes less restrictive assumptions about the smoothness of the underlying function for nonparametric part. The computational time is linear.
展开▼
