Working Paper Series, Center for Labor Studies,
Uppsala University, Department of Economics

Abstract: This paper considers the definition and identification of treatment effects on conditional transition probabilities. We show that even under sequential random assignment only the instantaneous average treatment effect is point identified. Because treated and control units drop out at different rates, randomization only ensures the comparability of treatment and controls at the time of randomization, so that long run average treatment effects are not point identified. Instead we derive informative bounds on these average treatment effects. Our bounds do not impose (semi)parametric restrictions, as e.g. proportional hazards, that would narrow the bounds or even allow for point identification. We also explore various assumptions such as monotone treatment response, common shocks and positively correlated outcomes.

Keywords: Partial identification; duration model; randomized experiment; treatment effect

JEL-codes: C14; C41

57 pages, March 18, 2011

Full text files

136494_20113.pdf  

Download statistics

• Uppsala University, Department of Economics
• Home page for this series

Questions (including download problems) about the papers in this series should be directed to Katarina Grönvall ()
Report other problems with accessing this service to Sune Karlsson ().

RePEc:hhs:uulswp:2011_003This page generated on 2024-09-13 22:17:29.