Asymptotic behavior of LSE estimator of an AR(1) coefficient with associated innovations

Abstract

This article is devoted to the asymptotic properties of the ordinary least squares estimator (LSE) of the coefficient ρ in a stationary autoregressive process of order one with negatively associated innovations. Under appropriate conditions, consistency and asymptotic normality are derived.

Keywords:

• Autoregressive process
• linear process
• negatively associated innovations
• sample covariance
• weak dependence

AMS CLASSIFICATION:

• Primary 62F12
• secondary 60F05