discrete choice modeling

This study investigates how methodological choices made by modelers during discrete choice model construction—particularly in noise pollution policy contexts—induce substantial variation in outcomes. Method: We develop the “Serious Choice Modeling Game,” a behavioral experiment platform that systematically tracks modelers’ data exploration, model specification, and interpretation processes, integrating operational logs, descriptive statistics, and multinomial logit modeling for quantitative analysis. Contribution/Results: We find that while data visualization is widespread, missing-data handling is frequently neglected; parsimony preferences and iteration quality significantly affect model fit and simplicity; and substantial heterogeneity exists across modeling strategies applied to identical data. Crucially, willingness-to-pay estimates vary markedly with methodological choices, undermining policy recommendation reliability. This work provides the first systematic, causal evidence linking modeler behavior to outcomes in policy-oriented choice models, offering empirical foundations for enhancing reproducibility and policy robustness of discrete choice analysis.

Conventional discrete choice models rely on manual trial-and-error and subjective assumptions, resulting in low efficiency and poor reproducibility; existing metaheuristic approaches treat model specification as a static optimization problem, ignoring historical estimation information and thus failing to enable dynamic search adaptation or cross-task knowledge transfer. Method: We propose the first deep reinforcement learning–based (DQN) automated model search framework, formalizing model specification as a sequential decision-making process. We design a reward function jointly optimizing goodness-of-fit and model parsimony, and employ a serialized structural encoding scheme. Contribution/Results: The method requires no domain-specific prior knowledge, supports dynamic exploration control and cross-scenario knowledge transfer, and consistently converges to high-quality models under diverse data-generating processes. It significantly improves search efficiency, robustness, and generalization capability compared to state-of-the-art alternatives.

Traditional discrete choice models (based on utility maximization) and cognitive process models—such as Decision Field Theory (DFT)—remain conceptually disjoint, and DFT’s process parameters lack empirical grounding in physiological evidence. Method: We propose a cognition-enhanced modeling framework that systematically integrates multimodal physiological signals—including eye-tracking, skin conductance, and heart rate—into DFT’s dynamic process parameters, thereby coupling discrete choice formalism with neurocognitive mechanisms. The framework is applied to both static accommodation choice and dynamic driving decision-making tasks. Results: In static choice, eye-tracking data significantly improves model interpretability and predictive accuracy. In dynamic driving, stress-related physiological measures synergize with oculomotor data to refine DFT parameter estimation, outperforming conventional fusion approaches. This work establishes an interpretable, empirically grounded bridge linking neural activity, cognitive dynamics, and observable behavior—advancing theory-driven, physiologically validated behavioral modeling.

This paper investigates the cross-decision independence of stochastic choice in diversified decision-making—specifically, whether choice probabilities remain invariant when irrelevant alternatives are added to the menu. We develop a menu-dependent stochastic choice model that captures endogenous randomness satisfying the “decomposability” axiom. Theoretically, we establish two key results: first, for general outcome spaces, decomposability is equivalent to the existence of a universal utility function under which choices follow a multinomial logit (MNL) rule; second, for monetary outcomes, it uniquely implies a one-parameter logit family. These characterizations are robust to approximate decomposability and label perturbations. Our framework unifies explanations of intertemporal choice, risky decision-making, and ambiguity preferences, providing the first axiomatized foundation for behavioral modeling grounded in stochastic choice theory.

This paper addresses identification and estimation in continuous-time dynamic discrete-choice games, focusing on the previously overlooked challenges of endogeneity and heterogeneity in decision arrival rates (i.e., timing of actions). Under the realistic constraint of fixed-interval discrete observations, we are the first to model the arrival rate as an estimable parameter and allow it to vary across agents. Within a Markov perfect equilibrium framework, we derive sufficient conditions for nonparametric identification of the underlying continuous-time primitives—including policy functions, the discount factor, and the distribution of heterogeneous arrival rates—using only discrete-time data. Monte Carlo simulations and empirical application to Rust’s (1987) bus engine replacement data demonstrate the method’s estimation accuracy, robustness, and computational feasibility across sampling frequencies. Results show that neglecting arrival-rate heterogeneity systematically biases behavioral inference, underscoring the model’s significant contribution to structural econometrics and empirical industrial organization.

This study presents the first systematic approach to inferring willingness-to-pay (WTP) from large language models (LLMs) in subjective choice scenarios lacking objectively correct answers. By presenting travel-assistance dilemmas and leveraging multinomial logit modeling, role-based prompting, and lightweight preference conditioning, the authors quantify the implicit WTP embedded in LLM responses and benchmark it against human data. The findings reveal that larger-scale LLMs can generate meaningful WTP estimates, yet they consistently exhibit attribute-level systematic biases, generally overestimating human WTP. Notably, incorporating minimal user preference conditioning substantially improves the alignment between LLM-derived valuations and actual human behavior, enhancing the behavioral fidelity of LLM-based economic inference.

This paper investigates the behavioral mechanisms underlying stochastic subset selection from a menu of alternatives. Addressing the proliferation of ad hoc functional forms and unclear behavioral foundations in existing menu-choice models, we develop a rigorous axiomatic framework that systematically characterizes the probabilistic structure of multi-item choice. Our method employs formal axiomatic analysis to derive necessary and sufficient conditions for key classes of parametric models. The main contributions are threefold: (i) a unified conceptualization of prominent models—including Random Utility, generalized Luce’s Axiom, and Sequential Elimination—clarifying their logical interconnections and fundamental distinctions; (ii) explicit identification of the behavioral assumptions embedded in each model and their domain of applicability; and (iii) derivation of empirically testable, high-discriminatory-power theoretical propositions. This framework advances the theoretical understanding of multi-item choice and provides a solid axiomatic foundation for empirical model testing and selection.

Existing decision theories lack a unified formal language, and key concepts—such as the distinction between subjective and objective perspectives and criteria for evaluating theoretical superiority—remain ambiguous, impeding rigorous comparison. This work addresses these limitations by constructing a unified decision-theoretic framework grounded in nonparametric structural equation models (NPSEMs), which precisely characterizes agents, causal relationships, and counterfactuals, and formally defines evidential decision theory (EDT) and causal decision theory (CDT). Building on this foundation, the paper proposes a personal decision theory that aims to maximize an individual’s subjective counterfactual utility and introduces a performance evaluation metric based on population-level interventions, proving its optimality under specific conditions. The framework enables formal analyses of Newcomb’s paradox and the smoking lesion problem, offering a clear modeling language and a comparable benchmark for decision theories.

This study addresses the limited theoretical grounding in contemporary traffic behavior modeling, which often relies heavily on data-driven AI predictions without adequately accounting for the motivational underpinnings of travel decisions. To bridge this gap, the paper introduces Goal Pursuit Theory (GPT) into transportation research for the first time, systematically incorporating multi-goal conflicts, context-dependent goal activation, and decision-making mechanisms across temporal scales. This approach transcends the constraints of traditional Random Utility Models (RUM) and Regret Minimization Models (RRM). By integrating hybrid choice modeling with matrix factorization techniques, the authors empirically demonstrate GPT’s superior explanatory power in contexts such as activity scheduling, vehicle ownership, and residential location choice, while also offering actionable implementation guidelines and benchmark data requirements.

This study addresses the computational inefficiency of dynamic discrete choice models in high-dimensional state-action spaces by uncovering an intrinsic connection between conditional choice simulation (CCS) and reinforcement learning (RL). The authors reformulate CCS estimation as an RL problem and propose a lightweight two-stage estimation framework that integrates RL’s path-based updating mechanism. This approach substantially improves computational efficiency while preserving the interpretability of structural models. Empirical evaluations in two canonical marketing contexts—machine replacement and consumer grocery purchasing—demonstrate that the proposed method significantly accelerates estimation compared to traditional CCS, thereby extending the applicability of structural models to high-dimensional, large-scale marketing problems.