A zero-inflated negative binomial regression model with hidden Markov chain

This paper proposes a zero-inflated negative binomial regression model with hidden Markov chain for panel count data with excess zeros, and provides an estimation method using the EM and quasi-Newton algorithms. An application to the analysis of FDI is given.

Saved in:
Bibliographic Details
Published inEconomics letters Vol. 92; no. 2; pp. 209 - 213
Main Authors Wang, Peiming, Alba, Joseph D.
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.08.2006
Elsevier
SeriesEconomics Letters
Subjects
Foreign direct investment
Foreign direct investment (FDI)
Markov chain
Markovian processes
Negative binomial distribution
Panel count data with excess zeros
Panel data
Serial dependence
Markov chain
C35
Panel count data with excess zeros
G21
F36
Foreign direct investment (FDI)
Serial dependence
Negative binomial distribution
Online AccessGet full text

Cover

More Information
Summary:This paper proposes a zero-inflated negative binomial regression model with hidden Markov chain for panel count data with excess zeros, and provides an estimation method using the EM and quasi-Newton algorithms. An application to the analysis of FDI is given.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0165-1765
1873-7374
DOI:10.1016/j.econlet.2006.02.009