Representing tourists’ heterogeneous choices of destination and travel party with an integrated latent class and nested logit model

Aiming at a better understanding of heterogeneous interdependencies between destination and travel party choices in tourism, this study attempts to simultaneously represent these two choices by integrating the nested logit model with the latent class modeling approach to accommodate both types of nested model structures together. Empirical analysis confirmed the effectiveness of the developed model, using a data collected from more than 2000 tourists in Japan. It was observed that on average the two types of nested model structures are almost equally shared by samples and the model structures could significantly vary with income level and gender. Influential factors related to choices of destination and travel party were also explored. Concretely speaking, travel time, attractiveness of destination and number of tourism spots were found to be important influential factors in destination choice, and gender, age, marital status have important effects on travel party choice.