A Zero-inflated Negative Binomial Regression Model with Hidden Markov Chain

This paper proposes a regression model for analysis of panel count data with the presence of excess zeros relative to a negative binomial distribution, in which the frequency distribution of counts changes according to an underlying two-state Markov chain. Features of the proposed model and estimation method are discussed. An application to the analysis of foreign direct investments in the United States by Japanese firms is given

MoreLess

Year of publication:	2004-08-11
Authors:	Wang, Peiming
Institutions:	Econometric Society
Subject:	Negative binomial distribution \| Panel count data with excess zeros \| Markov chain

More details

Series:	Econometric Society 2004 Far Eastern Meetings.
Type of publication:	Book / Working Paper
Notes:	The text is part of a series Econometric Society Far Eastern Meetings 2004 Number 644
Classification:	C33 - Models with Panel Data ; F21 - International Investment; Long-Term Capital Movements
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005342331