Estimation of Limited Dependent Variable Habit Persistence Models on Panel Data with an Application to the Dynamics of Self-employment in the former East Germany
Series
Contributions to economic analysis
ISBN
0-444-81548-1
Type
book section
Date Issued
1993
Author(s)
Editor(s)
Bunzel, H.
Jensen, P.
Westergard-Nielsen, N.
Abstract
Not available in German. The paper suggests two multi-step estimation methods for dynamic limited dependent variable models with random effects on panel data. The estimated coefficients are consistent, in the sense that the number of individuals tends to infinity and the number of time periods is fixed. The error terms are assumed to be independently and identical distributed across individuals, but otherwise unrestricted. The estimators have the advantages that they are asymptotically efficient in their proposed class, and that the vast literature on parametric and semi-parametric estimators of limited dependent variable models for cross-sections as well as the resulting software can be utilised. The small sample behaviour of the methods is investigated using a Monte Carlo study. The preferred method is applied to the analysis of the dynamics of self-employment in the former East Germany after unification.
Language
English
HSG Classification
contribution to scientific community
Refereed
No
Book title
Panel Data and Labour Market Dynamics
Publisher
North-Holland
Publisher place
Amsterdam
Number
222
Start page
263
End page
283
Pages
21
Subject(s)
Eprints ID
15314