Activity Number:
|
310
- Advances and Novel Problems in Flexible Analysis of Clustered Data with Complex Structures
|
Type:
|
Invited
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Nonparametric Statistics
|
Abstract #322314
|
View Presentation
|
Title:
|
Analysis of Clustered Complex Design Data: Propensity Score Matching
|
Author(s):
|
Mi-Ok Kim*
|
Companies:
|
UCSF
|
Keywords:
|
matching ;
clustered design
|
Abstract:
|
Matching is a popular approach to reducing confounding in observational studies. Various matching techniques/designs are available. Most of them were developed for causal inference with independent data, and little is known about their application to clustered data. The clustered data structure questions whether and how the data structure needs to be accounted for in matching. Matched data may further complicate the analysis by requiring both the clustered data structure and the matching structure to be accounted for. We extend the counterfactual outcomes framework to clustered data setting and delineate when and how to match and account for both the data and the matching structure. Simulation studies are conducted for comparison, which mimics a real world setting of multi-center kidney transplant cohort study.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2017 program
|