Optimal Model Averaging for Semiparametric Partially Linear Models with Censored Data

In the past few decades, model averaging has received extensive attention, and been regarded as a feasible alternative to selection. However, this work is mainly based on parametric framework complete dataset. This paper develops frequentist model-averaging estimation for semiparametric partially linear models with censored responses. The nonparametric function approximated by B-spline, weights in estimator are picked up via minimizing leave-one-out cross-validation criterion. resulting proved be asymptotically optimal sense of achieving lowest possible squared error. A simulation study demonstrates that method superior traditional model-selection methods. Finally, an illustration, proposed procedure further applied analyze two real datasets.