A Monte-Carlo study for propensity score methods.
- Author:
Tao WANG
1
;
Takuhiro YAMAGUCHI
;
Yasuo OHASHI
;
Xue-shan FENG
;
Qing-wu JIANG
Author Information
- Publication Type:Journal Article
- MeSH: Algorithms; Bias; Confounding Factors (Epidemiology); Data Interpretation, Statistical; Humans; Models, Statistical; Monte Carlo Method; Research Design; Risk Adjustment
- From: Chinese Journal of Epidemiology 2005;26(6):458-462
- CountryChina
- Language:Chinese
-
Abstract:
OBJECTIVETo evaluate the statistical property of the estimators on exposure effect and value of propensity score methods in practical use.
METHODSSimulation data, propensity score methods (PSM) were conducted to assess bias and efficiency, under both with/without model misspecification conditions.
RESULTSUnder model misspecification conditions, propensity score methods had better robustness than model based methods.
CONCLUSIONFor large and complicated data, propensity score methods had more flexibility in practical use than model based methods.