Improved HAC covariance matrix estimation based on forecast errors
Journal
Economics Letters
Journal Issue
99
Pages
89-92
Date Issued
2008
Author(s)
Abstract
We propose computing HAC covariance matrix estimators based on one-step-ahead forecasting errors. It is shown that this estimator is consistent and has smaller bias than other HAC estimators. Moreover, the tests that rely on this estimator have more accurate sizes without sacrificing its power. © 2007 Elsevier B.V. All rights reserved.
Type
journal article