Communication-efficient distributed estimation of causal effects with high-dimensional data

3227 

Contributed Abstract 

Poster 

Xiaohan Wang (1), Jiayi Tong (2), Sida Peng (3), Yong Chen (4), Yang Ning (1)

(1) Cornell University, N/A, (2) N/A, N/A, (3) Microsoft Research, United States, (4) University of Pennsylvania, Perelman School of Medicine, N/A

Jiayi Tong  
N/A

Sida Peng  
Microsoft Research

Yong Chen  
University of Pennsylvania, Perelman School of Medicine

Yang Ning  
Cornell University

Xiaohan Wang  
Cornell University

Xiaohan Wang  
Cornell University

We propose a communication-efficient algorithm to estimate the average treatment effect (ATE), when the data are distributed across multiple sites and the number of covariates is possibly much larger than the sample size in each site. Our main idea is to calibrate the estimates of the propensity score and outcome models using some proper surrogate loss functions to approximately attain the desired covariate balancing property. We show that under possible model misspecification, our distributed covariate balancing propensity score estimator (disthdCBPS) can approximate the global estimator, obtained by pooling together the data from multiple sites, at a fast rate. Thus, our estimator remains consistent and asymptotically normal. In addition, when both the propensity score and the outcome models are correctly specified, the proposed estimator attains the semiparametric efficiency bound. We illustrate the empirical performance of the proposed method in both simulation and empirical studies.

Causal Inference|High-dimensional Statistics|Double robustness|Distributed inference|Communication efficiency|Likelihood approximation

Section on Statistics in Epidemiology

Causal Inference