Nonparametric Bootstrap for Quasi-Likelihood Ratio Tests

Abstract

We introduce a nonparametric block bootstrap approach for Quasi-Likelihood Ratio type tests of nonlinear restrictions. Our method applies to extremum estimators, such as quasi-maximum likelihood and generalized method of moments estimators. Unlike existing parametric bootstrap procedures for Quasi-Likelihood Ratio type tests, our procedure constructs bootstrap samples in a fully nonparametric way. We study the higher order properties of our nonparametric block bootstrap and show the asymptotic refinements implied with respect to the standard asymptotic theory. Our approach delivers the same higher order properties of the nonparametric block bootstrap methods introduced in Andrews (2002) and Kim (2003) in relation to Wald and Lagrange Multiplier tests, respectively. Monte Carlo simulations confirm the accuracy of our bootstrap procedure.

Keywords: Block Bootstrap, Quasi-Likelihood Ratio Tests, Asymptotic Refinements

JEL Classification: C12, C13, C15