Statistical modelling of consumer behaviour

Market Research

Members of Smu have been heavily involved into several research projects on statistical models of market research. The projects have been supported companies such as Procter and Gamble and ACNielsen. The main research activity was concentrated around the Mixed Poisson model for describing consumer purchase behaviour (such as purchase incidence and brand choice) in markets.

Analysis of customer loyalty

We have also investigated the loyalty patterns of customers which are characterized by how often the customers change the preferred brands. The underlying idea of the model is the fact that customers change their loyalties not only under the influence of advertisement but also completely at random. We have found that switching behaviour heavily depends on the category (or even sub-category) and the brand equity and that this switching behaviour is rather stable in time.

Mixed Poisson models

We have done a large study which demonstrates the applicability of Mixed Poisson models for describing consumer purchase behaviour (such as purchase incidence and brand choice) in markets. Research in the present direction includes the numerical and theoretical study of dynamical models leading to these schemes, the study of probability distributions arising under different conditioning arguments and asymptotic properties of various statistical procedures concerning the parameters of the negative binomial and Dirichlet distributions. A large consumer panel containing the data about all purchase occasions of about 50000 households for 3 years is the main data set we use to test our models.

Cardiff Investigator

Collaborators

Dr. Vippal Savani, Dr. Nina Golyandina, Dr. Vladimir Nekrutkin.

Selected publications

  1. Savani, V, Zhigljavsky, A.A. (2006) Efficient estimation of parameters of the negative binomial distribution. Comm. Statist.: Theory and Methods 35, No. 4-6, 767--783.
  2. Savani, V., Zhigljavsky, A.A. (2007) Efficient parameter estimation for independent and INAR(1) negative binomial samples. Metrika, v. 65, no. 2, 207-255.
  3. Savani V., Zhigljavsky A. (2007) Asymptotic distributions of statistics and parameter estimates for mixed Poisson processes, Journal of Statist. Planning and Inference, v. 137, No. 12, 3990-4002.