AN EMPIRICAL-EVALUATION OF PARAMETER SENSITIVITY TO CHOICE SET DEFINITION IN SHOPPING DESTINATION CHOICE MODELS

This paper empirically examines parameter sensitivity to choice set sp ecification in the context of shopping destination choice, using super market choice data from Gainesville, Florida. We estimate parameters o f the widely applied multinomial legit (MNL) discrete choice model mul tiple times. Each estimation uses, for all observations, a single rand omly selected subset of the universal choice set. The distribution of parameter estimates is examined for specific market segments and choic e subset sizes. The results indicate that the parameters of the model can be quite sensitive to the selection of the choice set used in the calibration. However, this sensitivity is not even across all paramete rs and there are some interesting variations. Distance deterrence and chain image parameters, for example, exhibit much more stability than parameters for store size and store competition. In addition, model pa rameters show encouraging stability with relatively small choice sets of seven to ten stores.