JSM 2012 Home

JSM 2012 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Online Program Home

Abstract Details

Activity Number: 526
Type: Contributed
Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract - #306096
Title: A Forest Approach to Defining a Study Population
Author(s): Emil Pitkin*+ and Justin Bleich
Companies: The Wharton School and The Wharton School
Address: 4244 Spruce St., Philadelphia, PA, 19104, United States
Keywords: Observational Study ; Propensity Score ; Overlap ; Forest ; Classification

Following the seminal work of Rosenbaum and Rubin (1983), matching based on the propensity score has become the predominant matching technique in observational studies. In order to reduce the asymptotic variance of the estimated average treatment effect (ATE), overlap between the propensity score distributions of the treatment and control groups is desirable. Tree-based approaches describe the overlapping populations in terms of their covariates, rather than propensity scores. We develop a greedy algorithm, which, as in Small and Traskin (2011), classifies a study population with trees, but which directly reduces asymptotic variance of the ATE at each split. The method is compared to existing techniques on the NSW job training data.

The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2012 program

2012 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.