Modelling the effect of purchase quantity on consumer choice of product assortment

The authors develop a model to predict consumer selection of product assortment and its relationship to category purchase quantity. Brand and variety choice decisions are modelled with a nested logit. A shopper's vector of choice probabilities—and hence likely assortment selection—directly depends on his or her purchase quantity. The purchase quantity decision is modelled with a Poisson regression which is, in turn, integrated with the nested logit. The full model permits investigators to assess the impact that quantity—and the marketing activity that affects it—has on the assortment of items chosen at the point of purchase. It should provide a decision aid for both manufacturers and retailers seeking to understand and forecast demand for item assortment. The authors illustrate their modelling approach with scanner panel data for yogurt. The results show that panellists' assortment selection is significantly related to purchase quantity, but the overall magnitude of the effect is small when compared to other factors that predict item sales. © 1998 John Wiley & Sons, Ltd.