Equidispersion and moment conditions for count panel data model

This paper proposes some new moment conditions under the assumption of the equidispersion in count panel data model. These are obtained by using the association between variances and covariances in the disturbance. Some Monte Carlo experiments configured for the Poisson model show that the GMM estimators using the new moment conditions perform better than the conventional quasi-differenced GMM estimator and some gains are recognized in using the new moment conditions.

MoreLess

Year of publication:	2009-06
Authors:	Kitazawa, Yoshitsugu
Institutions:	Faculty of Economics, Kyushu Sangyo University
Subject:	count panel data \| linear feedback model \| equidispersion \| implicit operation \| crosslinkage moment conditions \| GMM \| Monte Carlo experiments

More details

Extent:	application/pdf
Series:	Discussion Papers.
Type of publication:	Book / Working Paper
Language:	English
Notes:	Number 33 21 pages
Classification:	C23 - Models with Panel Data
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005036777