Endogenous Attrition in Panels

Laurent Davezies and Xavier D'Haultfoeuille

No 2013-17, Working Papers from Center for Research in Economics and Statistics

Abstract: We consider endogenous attrition in panels where the probability of attrition may depend on current and past outcomes. We show that this probability is nonparametrically identified provided that instruments affecting the outcomes but not directly attrition, and whose distribution is identified, are available. We thus complement Hirano et al. (2001)’s framework, which does not rely on such instruments. Contrary to their approach, neither a refreshment sample nor an additive decomposition on the probability of attrition are needed. We also show that the exclusion restriction has testable implications. We propose an efficient estimation and a test of the exclusion restriction when the outcome and instruments are discrete. The continuous case, which shares some similar features with nonparametric instrumental variable additive models, is also investigated. Finally, we apply our results to the French labor force survey, and provide evidence that attrition is related to transitions on the employment status

Keywords: Panel data; Endogenous attrition; Instrumental variables (search for similar items in EconPapers)
JEL-codes: C14 C21 C25 (search for similar items in EconPapers)
Pages: 38
Date: 2013-10
New Economics Papers: this item is included in nep-ecm
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://crest.science/RePEc/wpstorage/2013-17.pdf Crest working paper version (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:crs:wpaper:2013-17

Access Statistics for this paper

More papers in Working Papers from Center for Research in Economics and Statistics Contact information at EDIRC.
Bibliographic data for series maintained by Secretariat General () and Murielle Jules Maintainer-Email : murielle.jules@ensae.Fr.