Abstract
This paper proposes a new, more general model of consumer brand choice behavior. The model is a composite of the Bernoulli, Markov, and Linear Learning Models, all of which have previously been proposed to represent brand choice behavior. The properties of the model are explored, and a parameter estimation technique is developed. A special case of the model is developed for situations in which insufficient data are available to estimate the parameters of the complete model.

This publication has 0 references indexed in Scilit: