Robust estimators for nearly nonstationary AR(1) models

Abstract

In this paper we consider the asymptotic behaviour of robust estimators with a high breakdown point and high efficiency for nearly nonstationary AR(1) processes. The asymptotic distribution of these estimators is derived. Simulated percentiles of these distributions are given. We also perform a Monte Carlo study to compare the robustness properties of these estimators with the least squares and M-estimators. The Monte Carlo results suggest that the estimators investigated possess good robustness properties over the least squares and M-estimators.