Benefit of adding an alternative to one׳s choice set: A regret minimization perspective

The aim of this paper is to present and test a crucial building block for a regret minimization based choice set formation model, in that it presents a regret based benefit measure for the value associated with adding an alternative to one׳s choice set. By doing so, the paper contributes to existing research which has predominantly adopted a utility-based perspective. I show, using simulations based on a route choice model estimated on stated choice data, that the two perspectives - regret based and utility based - generate markedly different benefits. These are caused by the fact that the regret based perspective takes into account choice set composition effects. For example, in line with its behavioral premises, the regret based model predicts that adding an attractive alternative to the set choice only results in a large reduction in regret when the alternative outperforms existing alternatives in terms of every attribute (i.e., becomes a 'clear winner'). In general, the benefit of adding a new alternative to one׳s choice set is predicted to be substantially higher by a utility based model, compared to a regret based counterpart. This implies that, to the extent that regret minimization (utility maximization) is an important determinant of decision-making, a utility (regret) based model would overestimate (underestimate) the true size of the decision-maker׳s choice set.