The Focal Multinomial Logit Model : Threshold Effects on Consumer Choice, Assortment, Pricing and Estimation

{Problem Definition:} This paper considers the operational management problems under a newly proposed choice model that captures the effect of focality. The offered assortment is separated into the focal set and the non-focal set under this new model due to the bias of focality, which is identified by the focal sets and an assortment-dependent focal parameter. A prospective consumer is more likely to choose a product from the focal set, while she may still choose one from the non-focal set for a variety of reasons such as previous purchase experience or brand loyalty. This focal multinomial logit model generalizes the famous multinomial logit model and several well-studied consideration-set choice models. In addition, it has the capability to describe and explain a variety of irrational choice behaviors often observed in practice, such as the context effect, halo effect, and choice overload. {Methodology/results:} In this paper, we primarily focus on the threshold focal set and various focal parameter settings, including the constant, cardinality, and linear focal multinomial logit models, as well as a broader model that satisfies certain regularity conditions and subsumes the above models. We analyze the computational complexity and propose polynomial-time exact or approximation algorithms to solve the assortment optimization problems under different focal parameters. We then characterize the optimal strategy for the joint price and assortment optimization problem. Additionally, we develop a mixed integer conic programming reformulation method that converges to a global optimal estimator for the model calibration problem. {Managerial Implications:} We use these methods to conduct numerical experiments on both synthetic and real data sets. The results demonstrate the efficiency of our proposed algorithms, the predictive power, and the increase in revenue for the focal multinomial logit model. Our extensive analysis implies that in practice retailers may take into account the effect of focality in consumer purchase behavior because it could increase the accuracy of demand estimation and therefore improve operational performance