Asymptotic properties of estimators in heteroscedastic partially linear EV models with ANA errors

Abstract

This article addresses the estimation problem associated with heteroscedastic partially linear error-in-variables (EV) models. We establish some strong consistency properties for estimators of both the slope parameter and the nonparametric component based on asymptotic negative association (ANA or ρ−, for short) errors when the error variance is known. In cases where the error variance is unknown, we further investigate the strong consistency properties for estimators of the slope parameter, nonparametric component, and variance function based on ANA samples. Additionally, simulations are conducted to evaluate the finite sample performance of these estimators and to validate our theoretical findings.