|
Activity Number:
|
141
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #305716 |
|
Title:
|
Efficient Nonparametric Estimation of Causal Effects in Randomized Trials with Noncompliance
|
|
Author(s):
|
Jing Cheng*+ and Dylan S. Small and Thomas R. Ten Have and Zhiqiang Tan
|
|
Companies:
|
University of Pennsylvania and University of Pennsylvania and University of Pennsylvania and Johns Hopkins University
|
|
Address:
|
423 Guardian Drive, 503 Blockley Hall, Biostatistics, Philadelphia, PA, 19104,
|
|
Keywords:
|
causal effects ; randomized trials ; noncompliance ; efficient nonparametric estimation ; compliance class
|
|
Abstract:
|
Causal approaches based on the potential outcome framework provide a useful tool for addressing the noncompliance problems in randomized trials. Various estimators (e.g., instrumental variable (IV) estimator) have been proposed for causal effects of treatment. We propose a new estimator by applying the empirical likelihood with moment restrictions to the mixture outcome distributions. Simulation studies show this estimator is robust to parametric distribution assumptions and more efficient than the standard IV estimator. The method is applied to data from a randomized trial of an encouragement intervention to improve adherence to prescribed depression treatments among depressed elderly patients in primary care practices.
|
