GPMatch: A Bayesian causal inference approach using Gaussian process covariance function as a matching tool

Frontiers in Applied Mathematics and Statistics

Applied Mathematics and Statistics

Siva Sivaganisan¹  Chen Chen²  Bin Huang³  Jinzhong Liu⁴ 

[1] Department of Mathematical Sciences, University of Cincinnati, Cincinnati, OH, United States;Division of Biostatistics and Epidemiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States;Division of Biostatistics and Epidemiology, Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States;Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH, United States;Regeneron Pharmaceuticals, Basking Ridge, NJ, United States;

关键词: causal inference;    matching;    doubly robust (DR) estimator;    marginal structural model;    G-estimation;    real world data (RWD);   

DOI  :  10.3389/fams.2023.1122114

 received in 2022-12-12, accepted in 2023-02-09,  发布年份 2023

来源: Frontiers

【 摘 要 】

A Gaussian process (GP) covariance function is proposed as a matching tool for causal inference within a full Bayesian framework under relatively weaker causal assumptions. We demonstrate that matching can be accomplished by utilizing GP prior covariance function to define matching distance. The matching properties of GPMatch is presented analytically under the setting of categorical covariates. Under the conditions of either (1) GP mean function is correctly specified; or (2) the GP covariance function is correctly specified, we suggest GPMatch possesses doubly robust properties asymptotically. Simulation studies were carried out without assuming any a priori knowledge of the functional forms of neither the outcome nor the treatment assignment. The results demonstrate that GPMatch enjoys well-calibrated frequentist properties and outperforms many widely used methods including Bayesian Additive Regression Trees. The case study compares the effectiveness of early aggressive use of biological medication in treating children with newly diagnosed Juvenile Idiopathic Arthritis, using data extracted from electronic medical records. Discussions and future directions are presented.

【 授权许可】

Unknown   
Copyright © 2023 Huang, Chen, Liu and Sivaganisan.