Towards a dynamic analysis of multiple-store shopping: evidence from Spanish panel data

Abstract

This paper aims to determine why consumer purchasing of fast moving consumer goods varies over time in Spain. More specifically, our objective is to explain multiple-store shopping in the households belonging to the Spanish Nielsen Homescan consumer panel that provides information about household shopping decisions between April 2003 and April 2004. In order to achieve this purpose, a Bayesian Dynamic Tobit model is used. The results allow us to confirm the influence of several demographical and geographical variables on household multiple-store shopping during the sample period.

Keyword : Bayesian inference, fast moving consumer goods (FMCG), MCMC, multiple-store shopping, purchasing patterns, store choice, Tobit model