"Heteroskedasticity and Autocorrelation Robust Inference: Recommendations for Practice"

at 3pm on Wednesday, October 21, 2015 in 310 Silsby

Please sign up for a meeting, lunch, or dinner at:

https://docs.google.com/spreadsheets/d/1l1wOV5GVU4wAyYOKaNSgY9bJSNiSHU-NWIZdsbFQ9D8/edit?usp=sharing

Abstract

The classic papers by Newey and West (1987) and Andrews (1991) spurred a large body of work investigating methods for improved heteroskedasticity- and autocorrelation-robust (HAR) inference in time series regression. Empirical practice, however, has not kept up with these developments. The goal of this paper is to draw on these developments to make a recommendation for practitioners about how to compute standard errors in typical regression and GMM settings with heteroskedasticity and serial correlation. A general conclusion from this literature is that the large size distortions of traditional HAR tests can be greatly reduced by using longer time-domain truncation parameters and modified “fixed-b” asymptotic critical values. In general, fixed-b critical values are nonstandard, however a class of tests (using basis-function estimators of the long-run variance) have standard t and F fixed-b asymptotic distributions. We provide new theoretical results on higher-order size distortions that permit ranking tests within this class, and we compare tests in extensive Monte Carlo experiments that reflect both the benchmark designs used in the literature and data-based designs using U.S. macroeconomic time series. Theory and Monte Carlo results suggest that little is lost by restricting attention to tests with fixed-b t- and F-distributions. We conclude that estimating the long-run variance by the equal-weighted periodogram, averaged over the first four periodogram ordinates, provides inference that is good in an absolute sense and represents a large improvement over currently-standard HAR methods.