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:
|
392
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, July 31, 2012 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract - #304761 |
Title:
|
Making Instrumental Variables Look More Like Experimental Design
|
Author(s):
|
Mike Baiocchi*+ and Dylan Small and Daniel Polsky
|
Companies:
|
Stanford University and The Wharton School and University of Pennsylvania
|
Address:
|
Department of Statistics, Stanford, CA, , United States
|
Keywords:
|
instrumental variables ;
causal inference ;
near/far matching ;
study design ;
propensity score ;
matching
|
Abstract:
|
Instrumental variable (IV) techniques are typically conceptualized and implemented under a structural equation modelling framework(e.g., two stage least squares). This is not the only framework for IV. A study design approach to IV has been developed. This technique, called near/far matching, follows the logic of a randomized controlled trial to obtain causal estimates in an observational setting.
We will start by introducing near/far matching and discussing its connection to propensity score matching. We will then introduce a methodology for designing a strong study using near/far matching, both when you believe you have a valid instrument and when you suspect a violation of the IV assumptions.
|
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.