Neural integrated choice model with ParetoTail and Gaussian copula for travel behavior analysis

Abstract

Travel behavior modeling is essential for transportation demand analysis and policy-making, yet traditional discrete choice models often struggle with real-world data complexities, such as heavy-tailed distributions and strong feature correlations. This study proposes a novel neural network framework integrated with advanced statistical techniques to effectively address these issues. Specifically, a ParetoTail transformation is employed to normalize heavy-tailed travel attributes, such as travel time and cost, reducing the undue influence of extreme values. To explicitly capture complex dependencies among features, a Gaussian copula approach is integrated, improving the robustness of the model against traditional independence assumptions. Furthermore, a gating mechanism is introduced to dynamically balance the contributions of continuous and discrete features, incorporating random noise to account for preference heterogeneity across individual travelers. Extensive empirical analyses, initially on the Swissmetro dataset and validated in three additional diverse public datasets, demonstrate that the proposed model consistently and significantly outperforms the baseline models (MNL, MXL, L-MNL, E-MNL, EL-MNL) in terms of prediction accuracy, F1 score, and AUC values. Crucially, the interpretability of the model reveals nuanced behavioral insights, such as the heterogeneity of decision-making styles across the population and non-linear responses to cost in long-distance travel. Additional ablation studies underscore the essential roles of the ParetoTail, Gaussian copula, and gating components. In general, this integrated framework provides a flexible, robust, and generalizable approach to modeling travel behavior, offering transport planners a more accurate tool for policy evaluation in complex real-world scenarios.