Asymptotic efficiency of the two-stage estimation method for copula-based models

Subscribe to email list

Please select the email list(s) to which you wish to subscribe.

You are here

Asymptotic efficiency of the two-stage estimation method for copula-based models

TitleAsymptotic efficiency of the two-stage estimation method for copula-based models
Publication TypeJournal Article
Year of Publication2005
AuthorsJoe, H
JournalJournal of Multivariate Analysis
Volume94
Pagination401-419
Date PublishedJUN
ISSN0047-259X
AbstractFor multivariate copula-based models for which maximum likelihood is computationally difficult, a two-stage estimation procedure has been proposed previously; the first stage involves maximum likelihood from univariate margins, and the second stage involves maximum likelihood of the dependence parameters with the univariate parameters held fixed from the first stage. Using the theory of inference functions, a partitioned matrix in a form amenable to analysis is obtained for the asymptotic covariance matrix of the two-stage estimator. The asymptotic relative efficiency of the two-stage estimation procedure compared with maximum likelihood estimation is studied. Analysis of the limiting cases of the independence copula and Frechet upper bound help to determine common patterns in the efficiency as the dependence in the model increases. For the Frechet upper bound, the two-stage estimation procedure can sometimes be equivalent to maximum likelihood estimation for the univariate parameters. Numerical results are shown for some models, including multivariate ordinal probit and bivariate extreme value distributions, to indicate the typical level of asymptotic efficiency for discrete and continuous data. (c) 2004 Elsevier Inc. All rights reserved.
DOI10.1016/j.jmva.2004.06.003