Deciding where to shop: Disaggregate random utility destination choice modeling of grocery shopping in canton Zurich

Abstract

The present work aims at deepening our understanding of the choice process of individuals deciding where to shop for groceries. Two main datasets were used: a trip dataset from the Microcensus for Swiss travel behavior 2005 in which all locations were geocoded; and a store attribute dataset compiled by the author from various sources. Choice sets were generated based to the travel time budget of individuals and a random sample with a fixed number of alternatives was drawn for the discrete choice modeling process. MNL models were run for trips by car and by walking. For the car model, the most relevant store attributes in the destination choice process were found to be the store size and distance required to reach the store in home and non-home based trips. Regarding socio-demographic characteristics of the individuals, income and household size were found to drive the decision over age and gender.