Leveraging Causal Graphs for Blocking in Randomized Experiments
Abstract: \textit{Randomized experiments} are often performed to study the causal effects of interest. \textit{Blocking} is a technique to precisely estimate the causal effects when the experimental material is not homogeneous. It involves stratifying the available experimental material based on the covariates causing non-homogeneity and then randomizing the treatment within those strata (known as \textit{blocks}). This eliminates the unwanted effect of the covariates on the causal effects of interest. We investigate the problem of finding a \textit{stable} set of covariates to be used to form blocks, that minimizes the variance of the causal effect estimates. Using the underlying causal graph, we provide an efficient algorithm to obtain such a set for a general \textit{semi-Markovian} causal model.
Submission Length: Regular submission (no more than 12 pages of main content)
Assigned Action Editor: ~Novi_Quadrianto1
Submission Number: 90
Loading