A1158
Title: Non-parametric variance function estimation in linear regression models with many regressors
Authors: Weilun Zhou - University of Cambridge (United Kingdom) [presenting]
Abstract: The nonparametric estimation of unknown variance function in linear regression models that allow for many regressors is considered. When the number of regressors increases at the same rate as the sample size, the conventional nonparametric variance estimator is biased. An orthogonal series expansion is used to approximate the unknown variance function and a leave-one-out slope coefficient estimator is involved to eliminate the bias. The asymptotic properties of the proposed estimator are derived. Simulation evidence consistent with the theoretical results and an empirical illustration is also provided.
