Simplifying Probabilistic Expressions in Causal Inference
Tikka, S., & Karvanen, J. (2017). Simplifying Probabilistic Expressions in Causal Inference. Journal of Machine Learning Research, 18, 1-30. Retrieved from http://jmlr.csail.mit.edu/papers/volume18/16-166/16-166.pdf
Published inJournal of Machine Learning Research
© 2017 Santtu Tikka and Juha Karvanen. This is an open access article distributed under the terms of a Creative Commons License.
Obtaining a non-parametric expression for an interventional distribution is one of the most fundamental tasks in causal inference. Such an expression can be obtained for an identifiable causal effect by an algorithm or by manual application of do-calculus. Often we are left with a complicated expression which can lead to biased or inefficient estimates when missing data or measurement errors are involved. We present an automatic simplification algorithm that seeks to eliminate symbolically unnecessary variables from these expressions by taking advantage of the structure of the underlying graphical model. Our method is applicable to all causal effect formulas and is readily available in the R package causaleffect.