“(Un) Expected consequences of becoming a new format shopper: A causal approach” joint with Paul Ellickson and Mitch Lovett, 2021
We quantify the causal impact of new format/platform entry on a broad range of consumer shopping behaviors, including overall spending, frequency of visit, breadth and depth of basket and brand loyalty. This is a large undertaking, made feasible only by recent innovations in data availability and statistical methodology. In particular, we exploit machine learning and data fusion techniques to combine a rich collection of data on consumer choice and store entry decisions into a unified longitudinal panel and employ and extend frontier synthetic control methods to address issues of selection and heterogeneous response inherent to this context.
Understanding the impact of biased social media on political races joint with Mitch Lovett and Paulo Albuquerque, 2020
Learning preferences for stores (2022)