Identification in Nonlinear Difference in Difference Models with Multivalued Treatment Outcomes

36 Pages Posted: 5 Nov 2011

See all articles by Carlos Cañón

Carlos Cañón

University of Toulouse 1 - Toulouse School of Economics (TSE)

Date Written: November 4, 2011

Abstract

We study the conditions to directly identify the joint distribution of outcomes for the treated group in absence of any treatment, avoiding to make assumptions that allow identify each counterfactual marginal distribution. Our starting point is Athey & Imbens (2006)'s Changes-In-Changes Model, but we generalize it letting the treatment also affect the distribution of unobservables even within each group (e.g. treated and untreated). We show that under a reasonable set of assumptions we can identify sharp bound for the counterfactual joint distribution of outcome variables. Moreover, we show identification power increases for copulas from the Archimedean family.

Keywords: treatment effect, identification, nonlinear difference in difference, copulas

JEL Classification: C14, C31, C46

Suggested Citation

Cañon, Carlos Ivan, Identification in Nonlinear Difference in Difference Models with Multivalued Treatment Outcomes (November 4, 2011). Available at SSRN: https://ssrn.com/abstract=1954587 or http://dx.doi.org/10.2139/ssrn.1954587

Carlos Ivan Cañon (Contact Author)

University of Toulouse 1 - Toulouse School of Economics (TSE) ( email )

31 Allee de Brienne
Bureau MF003
Toulouse, 31000
France
+33 668483626 (Phone)

HOME PAGE: http://canoncic.freeshell.org

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
69
Abstract Views
587
Rank
599,109
PlumX Metrics