Multiple Discoveries in Causal Inference: LATE for the Party

Abstract

Multiple discoveries in statistics may be more common than many readers realize. We tell a story of multiple discoveries in causal inference related to a quantity called the local average treatment effect (LATE). The development of LATE by Imbens and Angrist contributed greatly to their winning the 2021 Nobel Prize in Economics. We discuss how multiple investigators were either early or late for the “party” of intermediate and final discoveries involving LATE.

Additional information

Notes on contributors

Stuart G. Baker

Stuart G. Baker is a mathematical statistician in the National Cancer Institute Division of Cancer Prevention. He developed surrogate endpoint criteria, test tradeoff curves for evaluating risk prediction, the multinomial-Poisson transformation, the transitive fallacy plot, and the B-K plot for Simpson’s paradox. He also wrote papers on cancer biology paradoxes and proposed the detached pericyte hypothesis for tumorigenesis. He is a fellow of the American Statistical Association and was the first recipient of the distinguished alumni award from the Harvard Department of Biostatistics.

Karen S. Lindeman

Karen S. Lindeman is an associate professor in the Johns Hopkins University Department of Anesthesiology and Critical Care Medicine. She specializes in obstetrical anesthesia and serves in both clinical and educational capacities, instructing fellows, residents, and medical students. Her prior research includes basic science studies investigating mechanisms of airway and uterine muscle relaxation.