Controlling the Size of Autocorrelation Robust Tests

TL;DR

This paper explores how to control the size of autocorrelation robust tests by selecting suitable critical values to address their common size distortions and power issues.

Contribution

It provides conditions under which the size of autocorrelation robust tests can be effectively controlled through critical value adjustments.

Findings

Identifies conditions for size control of autocorrelation robust tests

Offers guidelines for choosing critical values to improve test reliability

Addresses the balance between size accuracy and test power

Abstract

Autocorrelation robust tests are notorious for suffering from size distortions and power problems. We investigate under which conditions the size of autocorrelation robust tests can be controlled by an appropriate choice of critical value.