Sufficient covariates and linear propensity analysis

Working within the decision-theoretic framework for causal inference, we study the properties of "sufficient covariates", which support causal inference from observational data, and possibilities for their reduction. In particular we illustrate the r^ole of a propensity variable by means of a simple model, and explain why such a reduction typically does not increase (and may reduce) estimation efficiency.

• Type: Journal
• Year: 2010
• Where: JMLR
• Authors: Hui Guo, A. Philip Dawid
• Keywords: Causal Inference, Decision-theoretic Framework, JMLR 2010, Sufficient Covariates