Learning Adjustment Sets from Observational and Limited Experimental Data

Estimating causal effects from observational data is not always possible due to confounding. Identifying a set of appropriate covariates (adjustment set) and adjusting for their influence can remove confounding bias; however, such often identifiable alone. Experimental allow unbiased effect estimation, but are typically limited in sample size therefore yield estimates high variance. Moreover, experiments performed on different (specialized) population than the interest. In this work, we introduce method that combines large experimental identify adjustment sets improve estimation target population. The scores an by calculating marginal likelihood given observationally-derived estimate, using putative set. make inferences constraint-based methods. We show additional when compared state-of-the-art