Multicell Experiments for Marginal Treatment Effect Estimation of Digital Ads

The authors thank the editor, Jean-Pierre Dubé; the associate editor; and three anonymous reviewers for detailed comments and feedback as well as Ivan Canay, Gastón Illanes, Garrett Johnson, TI Kim, and Ilya Morozov as well as seminar participants at Michigan Ross, MIT Sloan, the 2023 Bass FORMS Conference, the 2023 Marketing Science Conference, the 2023 Workshop on Institutions, Economic Behaviour and Economic Outcomes, the 2023 12th Triennial Choice Symposium, and the 2023 Annual MIT Conference on Digital Experimentation for helpful comments. B.R. Gordon holds concurrent appointments at Northwestern and as an Amazon Scholar. This paper describes work performed at Northwestern and is not associated with Amazon. This paper was previously circulated under the title “A multicell experimental design to recover policy relevant treatment effects with an application to online advertising.”