Conditional Heteroskedasticity in Count Data Regression: Self-Feeding Activity in Fish

The paper introduces a new approach to incorporating time dependent overdispersion for Poisson related regression models. To handle the added flexibility in conditional heteroskedasticity in time series count data some wellknown estimators are adapted and a GMM type estimator is suggested. The estimators are applied to a time series of self-feeding activity in Arctic charr. There is strong support for both a dynamic conditional mean function and a dynamic model for the overdispersion.

MoreLess

Year of publication:	2002-11-21
Authors:	Brännäs, Kurt ; Brännäs, Eva
Institutions:	Institutionen för Nationalekonomi, Umeå Universitet
Subject:	Poisson \| Overdispersion \| ARCH \| Estimation \| Self-Feeding \| Arctic Charr

More details

Extent:	application/pdf
Series:	Umeå Economic Studies.
Type of publication:	Book / Working Paper
Notes:	Published in Communications in Statistics: Theory and Methods, 2004, pages 2745-2758. The text is part of a series Umeå Economic Studies Number 595 13 pages
Classification:	C13 - Estimation ; C22 - Time-Series Models ; C25 - Discrete Regression and Qualitative Choice Models ; C51 - Model Construction and Estimation
Source:	RePEc - Research Papers in Economics

Persistent link: https://ebvufind01.dmz1.zbw.eu/10005651943