MODELING STORE CHOICES WITH CROSS-SECTIONAL AND POOLED CROSS-SECTIONAL DATA - A COMPARISON

Contrary to many other types of spatial decisions, shopping destinatio n choice behavior is highly repetitive. For the practitioner looking f or good predictors of store patronage, for reliable marginal utility e stimates and reliable market share predictions, a central concern is w ith the type of data best suited to the research question, given the e xisting logistic and financial constraints. Different approaches can b e recognized in the literature in which conventional discrete choice m odels are applied to shopping destination choice problems. In this pap er, two of the most common practices are assessed and compared. First, the choice model is estimated with all choices of a relevant destinat ion observed during a certain period of time (pooled cross-sectional d ata). The alternative approach consists in an estimation with the choi ce of the destination where the majority of purchases takes place (cro ss-sectional data). In the particular data set employed here, no evide nce is found to support the idea that a multinomial legit model estima ted with cross-sectional data does not perform as well as a model esti mated with pooled cross-sectional data. Both models are found to be si milar in their ability to identity the main predictors of store choice . Models developed on either data sets have marginal utility estimates that exhibit no statistically significant differences. Finally, marke t share predictions derived from both models are not statistically dif ferent. It appears, therefore, that there is no need to collect repeat ed patronage data over an extended period of time. The practitioner wh o wishes to use a conventional discrete choice model may avoid spendin g much time and money by gathering limited data on regular patronage p atterns. In addition to this practical implication, the conclusions su ggest that regular shopping destinations are chosen in accordance with the same behavioral motives as ancillary destinations are.