On bandwidth selection in partial linear regression models under dependence

Abstract

We obtain the expression of an asymptotically optimal bandwidth for a semiparametric least-squares estimator of β in the model y= x T β+m(t)+ε , where x is random, t is fixed, m is unknown and ε is strong mixing. The selection method is based on second-order approximations for the variance and bias. Asymptotic normality is also established.