Publications
Publications
- April–June 2022
- INFORMS Journal on Data Science
Commentary on 'Causal Decision Making and Causal Effect Estimation Are Not the Same... and Why It Matters'
Abstract
There has been a substantial discussion in various methodological and applied literatures around causal inference; especially in the use of machine learning and statistical models to understand heterogeneity in treatment effects and to make optimal decision allocations. I commend Fernández-Loría and Provost (2021) for highlighting the important, and in retrospect, intuitive differences between causal estimation and causal decision making. This commentary is aimed at expanding the conversation in fruitful directions, with an eye toward real-world practice in organizations. Specifically, I highlight that future work will need to address how to exploit these theoretical results in practice (Section 2), how to ensure that causal decisions are fair (Section 3), and what are the additional benefits and challenges when there is a human in the loop (Section 4).
Keywords
Causal Inference; Treatment Effect Estimation; Treatment Assignment Policy; Human-in-the-loop; Decision Making; Fairness
Citation
McFowland III, Edward. "Commentary on 'Causal Decision Making and Causal Effect Estimation Are Not the Same... and Why It Matters'." INFORMS Journal on Data Science 1, no. 1 (April–June 2022): 21–22.