A0655
Title: Optimal model averaging for single-index models with divergent dimensions
Authors: Wendun Wang - Erasmus University Rotterdam (Netherlands) [presenting]
Abstract: A new approach is offered to address the model uncertainty in (potentially) divergent-dimensional single-index models (SIMs). A model-averaging estimator based on cross-validation, which allows the dimension of covariates, and the number of candidate models to increase with the sample size, is proposed. It is shown that when all candidate models are misspecified, the model-averaging estimator is asymptotically optimal, with its squared loss asymptotically identical to that of the infeasible best possible averaging estimator. In a different situation where correct models are available in the model set, the proposed method assigns all weights to the correct models asymptotically. Averaging regularized estimators and prescreening methods to deal with high-dimensional covariates are also proposed. The method via simulations and an empirical application are illustrated.
