|This centre is a member of The LSE Research Laboratory [RLAB]: CASE | CVER | CEP | FMG | SERC | STICERD||Cookies?|
Paper No' EM/2003/462: | Full paper
Save Reference as: BibTeX File | EndNote Import File
Keywords: Causality tests; long range, bootstrap tests.
Is hard copy/paper copy available? YES - Paper Copy Still In Print.
This Paper is published under the following series: Econometrics
Share this page: Google Bookmarks | Facebook | Twitter
Abstract:This paper examines a nonparametric test for Granger-causality for a vector covariance stationary linear process under, possibly, the presence of long-range dependence. We show that the test converges to a non-distribution free multivariate Gaussian process, say vec (B(μ)) indexed by μ Є [0,1]. Because, contrary to the scalar situation, it is not possible, except in very specific cases, to find a time transformation g(μ) such that vec (B(g(μ))) is a vector with independent Brownian motion components, it implies that inferences based on vec (B(μ)) will be difficult to implement. To circumvent this problem, we propose bootstrapping the test by two alternative, although similar, algorithms showing their validity and consistency.
Copyright © STICERD & LSE 2005 - 2015 | LSE, Houghton Street, London WC2A 2AE | Tel: +44(0)20 7955 6699 | Email: email@example.com | Site updated 23 May 2015