Working Paper
Product Choice with Large Assortments: A Scalable Deep-Learning Model
Dec 15, 2020
Develops a scalable product choice model that predicts customer-specific purchase probabilities for all products in an assortment in response to personalized discounts. The deep learning architecture (1) applies to raw transaction data from loyalty programs; (2) scales to large product assortments and customer bases; and (3) accounts for cross-product relationships and the effects of marketing interventions within and across categories.