{"title":"Limitations of the propensity scores approach: A simulation study","authors":"Igor Mandel","doi":"10.3233/mas-241505","DOIUrl":null,"url":null,"abstract":"Propensity scores (PS) have been studied for many years, mostly in the aspect of confounder matching in the control and treatment groups. This work is devoted to the problem of estimation of the causal impact of the treatment versus control data in observational studies, and it is based on the simulation of thousands of scenarios and the measurement of the causal outcome. The generated treatment effect was added in simulation to the outcome, then it was retrieved using the PS and regression estimations, and the results were compared with the original known in the simulation treatment values. It is shown that only rarely the propensity score can successfully solve the causality problem, and the regressions often outperform the PS estimations. The results support the old philosophical critique of the counterfactual theory of causation from a statistical point of view.","PeriodicalId":35000,"journal":{"name":"Model Assisted Statistics and Applications","volume":"44 22","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Model Assisted Statistics and Applications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3233/mas-241505","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Mathematics","Score":null,"Total":0}
引用次数: 0
Abstract
Propensity scores (PS) have been studied for many years, mostly in the aspect of confounder matching in the control and treatment groups. This work is devoted to the problem of estimation of the causal impact of the treatment versus control data in observational studies, and it is based on the simulation of thousands of scenarios and the measurement of the causal outcome. The generated treatment effect was added in simulation to the outcome, then it was retrieved using the PS and regression estimations, and the results were compared with the original known in the simulation treatment values. It is shown that only rarely the propensity score can successfully solve the causality problem, and the regressions often outperform the PS estimations. The results support the old philosophical critique of the counterfactual theory of causation from a statistical point of view.
期刊介绍：
Model Assisted Statistics and Applications is a peer reviewed international journal. Model Assisted Statistics means an improvement of inference and analysis by use of correlated information, or an underlying theoretical or design model. This might be the design, adjustment, estimation, or analytical phase of statistical project. This information may be survey generated or coming from an independent source. Original papers in the field of sampling theory, econometrics, time-series, design of experiments, and multivariate analysis will be preferred. Papers of both applied and theoretical topics are acceptable.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
