Estimating complex causal effects from incomplete observational data
This work provides a practical guide for statisticians to handle causal inference in challenging scenarios, but it is incremental as it relies on established methods without introducing new techniques.
The paper tackles the problem of estimating complex, highly nonlinear causal effects from observational data with missing values, demonstrating that a trained statistician can achieve this by combining existing tools like causal calculus and multiple imputation.
Despite the major advances taken in causal modeling, causality is still an unfamiliar topic for many statisticians. In this paper, it is demonstrated from the beginning to the end how causal effects can be estimated from observational data assuming that the causal structure is known. To make the problem more challenging, the causal effects are highly nonlinear and the data are missing at random. The tools used in the estimation include causal models with design, causal calculus, multiple imputation and generalized additive models. The main message is that a trained statistician can estimate causal effects by judiciously combining existing tools.