We’ve defined causal effects as an interventional distribution and posit two identification strategies to estimate them: the back-door and the front-door criteria. However, we cannot always use these criteria; sometimes, we cannot measure the necessary variables to use either of them.
More generally, given a causal model and some incomplete set of measurements, when is the causal effect of interest identifiable? In this blog post, we will develop a graphical criterion to answer this question by exploiting the concept of c-components. Finally, we will put the criterion in practice with multiple examples.