A unified approach to identifying, estimating and testing demand structures with aggregate scanner data

A one-to-one map between a specific demand structure and a restricted cross-elasticity matrix is developed and tested using a random utility model. The resulting demand structure can be used to evaluate a firm’s ability to differentiate its products, and can help determine if demand is characterized by factors that are easy to change. Similar brands are assumed to share common attributes that are stochastically evaluated by the consumer, resulting in a random utility model with correlated errors. The map results in more precise cross-elasticity estimates and can be used to both identify and test possible structures.