Going the extra mile: Estimating the willingness to travel to meet with friends using a joint destination choice model

Abstract

The purpose of this study is to develop a joint activity destination choice model to analyze the effect of clique-level characteristics on leisure behavior and evaluate the effect of incorporating clique-level attributes on the out-of-sample predictive ability of the model. To do so, we propose a discrete choice model that includes the preference for the ego and the cliques’ travel time; allowing us to estimate the willingness to travel to reduce the average clique travel time, showing that in joint destination choice there is a trade-off between the travel times of the individuals participating in the activity. We then compared the Clique model with an Ego model in terms of out-of-sample predictive power, showing that the Ego model overpredicts closer venues, underestimating the total kilometers traveled. The results show the importance of considering social networks in destination choice when such models are used for transport policy and agent-based simulations; if the models only consider the ego as the decision maker, these models will vastly underestimate the total kilometers traveled.