Structural models of reference price effects and brand equity

In this dissertation, I propose and estimate structural models for the reference-price phenomenon and for measuring brand equity.Chapters 2 and 3 show that a complete accounting of reference-price effects can only be done in a model where consumers are forward-looking and use past prices to form expectations of future prices.A number of behavioral theories argue that a consumer's response to a brand is not simply a function of its current price, but also a function of how that price compares to a reference price. A price higher than the reference price is perceived as a "loss," a price lower than the reference price is perceived as a "gain," and "losses" loom larger than "gains" (loss-aversion).Chapters 4 and 5 propose a structural method for measuring brand equity based on an equilibrium framework. In particular, brand equity is defined as excess profit due to the brand name, after controlling for observable attributes, and taking into account the impact of brand equity on the firm's equilibrium prices and marketing activities. A central feature of my methods is that I consider the equilibrium implications of a brand losing its equity. This equilibrium method gives better results than other methods including the revenue premium method because it gives managers richer and more intuitive information about brand equity by using a sophisticated model for marketing research. Furthermore, it builds a bridge between advanced marketing research techniques and real business decisions. (Abstract shortened by UMI.)In this study, both forward-looking behavior and unobserved heterogeneity are incorporated in a dynamic model of consumer choice, which is estimated on scanner panel data on yogurt and ketchup.