|
Activity Number:
|
361
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
WNAR
|
| Abstract - #308141 |
|
Title:
|
Calculating the Effect of a Randomized Intervention from Panel Data Subject to Noncompliance
|
|
Author(s):
|
Siddhartha Chib*+ and Liana Jacobi
|
|
Companies:
|
Washington University in St. Louis and The University of Melbourne
|
|
Address:
|
Olin School of Business, Campus Box 1133, Saint Louis, MO, 63130,
|
|
Keywords:
|
Bayesian inference ; Markov chain Monte Carlo ; treatment effect ; Potential outcomes ; predictive distributions
|
|
Abstract:
|
Motivated by the design of an actual experiment, we consider the problem of calculating the effect of a baseline randomized intervention from panel outcomes observed in subsequent periods. The main complication is that subject compliance with the intervention is less than perfect, perhaps due to observed and unobserved confounders. The goal of this study is to explore two alternative ways of addressing this complication in the context of a specific setting, the JOBS II intervention trial which was implemented to test whether an intervention in the form of a job training program can improve mental health and promote re-employment.
|
