Estimating a random-coefficients sample-selection model using generalized maximum entropy

We use the generalized maximum entropy (GME) method to estimate a sample-selection model and compare results with OLS and FIML. It is shown that GME outperforms FIML for small samples. In addition, we treat the heterogeneity of the units in the sample data by allowing coefficient vectors to deviate...

Full description

Saved in:
Bibliographic Details
Published in:Economics letters 2004-07, Vol.84 (1), p.87-92
Main Author: Peeters, Ludo M.K.
Format: Article
Language:English
Subjects:
Entropy
Estimation
Generalized maximum entropy
Methodology
Random coefficients
Sample-selection model
Samples
Small samples
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We use the generalized maximum entropy (GME) method to estimate a sample-selection model and compare results with OLS and FIML. It is shown that GME outperforms FIML for small samples. In addition, we treat the heterogeneity of the units in the sample data by allowing coefficient vectors to deviate randomly from a common sample mean.
ISSN:0165-1765
1873-7374
DOI:10.1016/j.econlet.2003.12.014