Bivariate Index Estimator for Evaluation
Type
presentation
Date Issued
1999-09-01
Author(s)
Froelich, Markus
Abstract
This paper develops a new estimator for microeconometric evaluation. This bivariate index estimator, which makes intensive use of bivariate nonparametric estimation on basis of two indices capturing the individual characteristics, is an extension of the matching approaches and the semiparametric sample selection models (or selectivity models), that dominate the evaluation literature and it's application. It unifies the virtues of both, allowing for selection on unobservables without parametrically specifying the outcome and selection functions. This paper derives the estimator and conducts a Monte Carlo study comparing it to its two main rivals. Download Manuscript: (pdf, 2228 kb)
Language
English
Keywords
matching
HSG Classification
contribution to scientific community
Refereed
No
Event Title
ESEM 99
Event Location
Santiago de Compostela
Subject(s)
Division(s)
Eprints ID
14939