The integer programing extreme value (IPEV) model: An application for estimation of the leisure trip demand

Abstract

We developed an integer programing extreme value (IPEV) model that accounts the integer property of trip data and has the same advantages as the multiple discrete–continuous extreme value choice (MDCEV) model. The proposed model is consistent with utility theory and provides a single structural framework for simultaneously modeling the choice of alternatives and quantity decisions with the constraint of the integer value of consumption. We demonstrate that the proposed model has a closed-form probability expression. Finally, we apply the proposed model to the recreation demand for national parks in Japan. The empirical results suggest that the proposed model provides a better fit for the data than the previous model and that ignoring the integer property of demand might cause an underestimation of the welfare loss.