Some Methods of Propensity‐Score Matching had Superior Performance to Others: Results of an Empirical Investigation and Monte Carlo simulations

Propensity‐score matching is increasingly being used to reduce the impact of treatment‐selection bias when estimating causal treatment effects using observational data. Several propensity‐score matching methods are currently employed in the medical literature: matching on the logit of the propensity...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Biometrical journal Ročník 51; číslo 1; s. 171 - 184
Hlavní autor:	Austin, Peter C.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Weinheim WILEY‐VCH Verlag 01.02.2009 Wiley-VCH
Témata:	Acute myocardial infarction Algorithms Applications Balance Bias Biology, psychology, social sciences Biometry - methods Canada - epidemiology Confounding Factors (Epidemiology) Data Interpretation, Statistical Distribution theory Exact sciences and technology General topics Humans Hydroxymethylglutaryl-CoA Reductase Inhibitors - therapeutic use Incidence Matching Mathematics Monte Carlo Method Monte Carlo simulations Myocardial Infarction - drug therapy Myocardial Infarction - mortality Numerical analysis Numerical analysis. Scientific computation Numerical methods in probability and statistics Observational studies Prescriptions - statistics & numerical data Probability and statistics Propensity score Propensity‐score matching Proportional Hazards Models Sciences and techniques of general use Statistics Survival Analysis Survival Rate Biometrics Rank statistic Statistical distribution Bias Large sample Stochastic method Economic sciences Parametric method Matching Propensity score Standard deviation Approximation theory Observational studies Propensity-score matching Monte Carlo method Matched sample Treatment efficiency Mortality Life science Statistical estimation Relative risk Statistical method Selection problem Numerical analysis Monte Carlo simulations Simulation Treatment effect Econometrics Acute myocardial infarction Balance Biased estimation
ISSN:	0323-3847, 1521-4036, 1521-4036
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Buďte první, kdo okomentuje tento záznam!