Influence of measurement error in discrete choice models: Utility maximizing versus random regret models

The aim of this study is to address the uncertainty problem caused by measurement error in random utility and random regret choice models. Based on formal analysis and empirical comparison, we provide new insights about the uncertainty problem in discrete choice modeling. Using standard assumptions, random measurement error is introduced into level-of-service variables. The effect of measurement error is analysed by comparing the estimated parameters of the concerned choice models, before and after introducing measurement error. We argue that although measurement error leads to biased estimation results in both types of models, bias appears differently in these choice models because random regret models involve a comparison of alternatives, and therefore uncertainty tends to accumulate. Therefore, bias tends to be larger. Moreover, since random regret models are constructed assuming semi-compensatory decision processes, uncertainty is not changed in the noncompensatory area. Several approaches are discussed to overcome this uncertainty problem.