Asymptotic normality for the estimators in heteroscedastic semiparametric EV models with α-mixing errors

Abstract In this paper, the heteroscedastic semi-parametric errors-in-variables (EV) model, yi = ξiβ + g(ti) + ϵi, xi = ξi + μi, is considered, where ϵi = σiei, σ2 i = f(ui), β is an unknown parameter to be estimated, g(·) and f(·) are unknown functions to be estimated. The asymptotic normalities for the estimators of β, g(·) and f(·) are presented under α-mixing random errors in either the errors are homoscedastic or heteroscedastic. Simulation studies are concerned to verify the finite sample behavior of the estimators. Mathematical Subject Classification: 62E20; 62F12; 62G20