A Simulation study of some Modern Weighting Methods for Estimating Treatment Effects in Observational Studies

Article Sidebar

Published Jul 17, 2020
DOI : https://doi.org/10.56801/Jmasm.V19.i2.4

Main Article Content

Lateef Babatunde Amusa
Department of Statistics, University of Ilorin, Ilorin, Nigeria
Temesgen Zewotir
Department of Statistics, School of Mathematics, Statistics and Computer Science, University of Kwazulu-Natal, Durban, South Africa
Delia North
Department of Statistics, School of Mathematics, Statistics and Computer Science, University of Kwazulu-Natal, Durban, South Africa

Abstract

Using extensive Monte Carlo simulations, we compared the performance of entropy balancing, empirical balancing calibration weighting (EBCW), covariate balancing propensity score, and the inverse probability treatment weighting methods,viathe performance of treatment effects estimation. The consistently superior performance of EBCW and entropy balancing leads us to recommend these two modernweighting techniques.

Article Details

Issue
Vol. 19 (2020): Issue 2
Section
Articles