The Error in Rejection Probability of Simple Autocorrelation Robust Tests

Econometrica · 2004 · Volume 72 · Issue 3

DOI: 10.1111/j.1468-0262.2004.00517.x

Abstract

A new class of autocorrelation robust test statistics is introduced. The class of tests generalizes the Kiefer, Vogelsang, and Bunzel (2000) test in a manner analogous to Anderson and Darling's (1952) generalization of the Cramér–von Mises goodness of fit test. In a Gaussian location model, the error in rejection probability of the new tests is found to beO(T-1logT), whereTdenotes the sample size.

Keywords

Asymptotic expansion, autocorrelation robust inference

Enable JavaScript for citation tools and additional paper interactions.