Publication:
Nonparametric Tests for Conditional Symmetry in Dynamic Models

Thumbnail Image
Identifiers
URI: https://hdl.handle.net/10016/2494
ISSN: 0304-4076
DOI: 10.1016/j.jeconom.2006.10.011
Files
NonparametricTest_JE_2007_ps.pdf (533.63 KB)
Publication date
2007
Defense date
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Elsevier
Impact
Google Scholar
Export
Research Projects
Organizational Units
Journal Issue
Abstract
This article proposes omnibus tests for conditional symmetry around a parametric function in a dynamic context. Conditional moments may not exist or may depend on the explanatory variables. Test statistics are suitable functionals of the empirical process of residuals and explanatory variables, whose limiting distribution under the null is nonpivotal. The tests are implemented with the assistance of a bootstrap method, which is justified assuming very mild regularity conditions on the specification of the center of symmetry and the underlying serial dependence structure. Finite sample properties are examined by means of a Monte Carlo experiment.
Description
Keywords
Omnibus tests, Symmetry tests, Conditional distributions, Time series, Empirical processes, Bootstrap
Bibliographic citation
Journal of Econometrics. 2007, vol. 141, p. 652-682
Collections
DE - Artículos de Revistas
Economists Online