Title: Distributed Design for Causal Inferences on Big Observational Data
Co-Authors: Arman Sabbaghi, Visiting Scholar in the Department of Statistics at the University of California Berkeley and Associate Professor in the Department of Statistics at Purdue University and Yumin Zhang, PhD Student, Department of Statistics, Purdue University
Abstract:
A fundamental issue in causal inference for Big Observational Data is confounding due to covariate imbalances between treatment groups. This issue can be addressed by designing data prior to analysis. However, existing design methods, developed for traditional observational studies with single designers, can yield unsatisfactory designs with suboptimum covariate balance for Big Observational Data due to their inability to accommodate the massive dimensionality, heterogeneity, and volume of the Big Data. We propose a new framework for the distributed design of Big Observational Data amongst collaborative designers. Our framework first assigns subsets of the high-dimensional and heterogeneous covariates to multiple designers. The designers then summarize their covariates into lower-dimensional quantities, share their summaries with the others, and design the study in parallel based on their assigned covariates and the summaries they receive. The final design is selected by comparing balance measures for all covariates across the candidates. We perform simulation studies and analyze thousands of datasets from the 2016 Atlantic Causal Inference Conference Data Challenge to demonstrate the power of our framework for yielding designs with good covariate balance for Big Observational Data. Our distributed design framework enables great flexibility and advantages for obtaining valid causal inferences from Big Observational Data.
*************************************************************
For those that can't attend in person:
Join Zoom Meeting
https://illinois.zoom.us/j/81579704788?pwd=ZHRwUU9rOHRHVklQOFB1cnRrMnQ5QT09
Meeting ID: 815 7970 4788
Password: 646606