INFORMATION-THEORY AS A UNIFYING STATISTICAL APPROACH FOR USE IN MARKETING-RESEARCH

Information theory is shown to provide a unified approach to a wide ra nge of problems in marketing research. For instance, this approach can be used to obtain characterizations parallel to those of the Hendry s ystem and other entropic approaches with great economy of assumptions and with the added flexibility that constraints can be easily identifi ed for explicit consideration and implemented as needed. Goodness-of-f it tests and decision modelling structures are supplied from these sam e stochastic models with a range of applications that include market s egmentation and brand shifting choices. A basic approach to these and other procedures is therefore obtainable from information theoretic me thods. These methods can be used to address stochastic model selection problems and other probabilistic models of marketing choice - for exa mple, Minimum Discrimination Information (MDI) estimation, Legit, Mult iplicative Competitive Interaction (MCI), and other important choice m odels are also shown in this paper to arise naturally from information theoretic formulations with duality relations developed by Charnes an d Cooper providing additional simplifications and interpretations.