|
on Discrete Choice Models |
By: | Hoang Giang Pham; Tien Mai; Minh Ha Hoang |
Abstract: | In this paper, we revisit parameter estimation for multinomial logit (MNL), nested logit (NL), and tree-nested logit (TNL) models through the framework of convex conic optimization. Traditional approaches typically solve the maximum likelihood estimation (MLE) problem using gradient-based methods, which are sensitive to step-size selection and initialization, and may therefore suffer from slow or unstable convergence. In contrast, we propose a novel estimation strategy that reformulates these models as conic optimization problems, enabling more robust and reliable estimation procedures. Specifically, we show that the MLE for MNL admits an equivalent exponential cone program (ECP). For NL and TNL, we prove that when the dissimilarity (scale) parameters are fixed, the estimation problem is convex and likewise reducible to an ECP. Leveraging these results, we design a two-stage procedure: an outer loop that updates the scale parameters and an inner loop that solves the ECP to update the utility coefficients. The inner problems are handled by interior-point methods with iteration counts that grow only logarithmically in the target accuracy, as implemented in off-the-shelf solvers (e.g., MOSEK). Extensive experiments across estimation instances of varying size show that our conic approach attains better MLE solutions, greater robustness to initialization, and substantial speedups compared to standard gradient-based MLE, particularly on large-scale instances with high-dimensional specifications and large choice sets. Our findings establish exponential cone programming as a practical and scalable alternative for estimating a broad class of discrete choice models. |
Date: | 2025–09 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.01562 |
By: | Jaap H. Abbring; {\O}ystein Daljord; Fedor Iskhakov |
Abstract: | We study the identification of dynamic discrete choice models with sophisticated, quasi-hyperbolic time preferences under exclusion restrictions. We consider both standard finite horizon problems and empirically useful infinite horizon ones, which we prove to always have solutions. We reduce identification to finding the present-bias and standard discount factors that solve a system of polynomial equations with coefficients determined by the data and use this to bound the cardinality of the identified set. The discount factors are usually identified, but hard to precisely estimate, because exclusion restrictions do not capture the defining feature of present bias, preference reversals, well. |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.07286 |
By: | Romuald Meango; Marc Henry; Ismael Mourifie |
Abstract: | Can stated preferences inform counterfactual analyses of actual choice? This research proposes a novel approach to researchers who have access to both stated choices in hypothetical scenarios and actual choices, matched or unmatched. The key idea is to use stated choices to identify the distribution of individual unobserved heterogeneity. If this unobserved heterogeneity is the source of endogeneity, the researcher can correct for its influence in a demand function estimation using actual choices and recover causal effects. Bounds on causal effects are derived in the case, where stated choice and actual choices are observed in unmatched data sets. These data combination bounds are of independent interest. We derive a valid bootstrap inference for the bounds and show its good performance in a simulation experiment. |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.13552 |
By: | Hung Tran; Tien Mai; Minh Ha Hoang |
Abstract: | The recursive logit (RL) model has become a widely used framework for route choice modeling, but it suffers from a key limitation: it assigns nonzero probabilities to all paths in the network, including those that are unrealistic, such as routes exceeding travel time deadlines or violating energy constraints. To address this gap, we propose a novel Constrained Recursive Logit (CRL) model that explicitly incorporates feasibility constraints into the RL framework. CRL retains the main advantages of RL-no path sampling and ease of prediction-but systematically excludes infeasible paths from the universal choice set. The model is inherently non-Markovian; to address this, we develop a tractable estimation approach based on extending the state space, which restores the Markov property and enables estimation using standard value iteration methods. We prove that our estimation method admits a unique solution under positive discrete costs and establish its equivalence to a multinomial logit model defined over restricted universal path choice sets. Empirical experiments on synthetic and real networks demonstrate that CRL improves behavioral realism and estimation stability, particularly in cyclic networks. |
Date: | 2025–09 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.01595 |
By: | Thimo De Schouwer; Elisabeth Gsottbauer; Iris Kesternich; Heiner Schumacher |
Abstract: | Work meaning can be an important driver of labor supply. Since, by definition, work meaning is associated with benefits for others, it also has an important fairness dimension. In a theoretical model, we show that workers’ willingness to pay for work meaning can be positive or negative, depending on the relative strength of fairness concerns and meaning preferences. To examine the importance of these behavioral motives for labor supply, we conduct a survey experiment with representative samples from the Netherlands and Germany in which we vary within-subject the benefits that a job creates for others. We find that only a minority of workers are actually willing to sacrifice wage for work meaning. The average willingness to pay for work meaning is positive, but substantially lower than the willingness to pay for job flexibility. There is a strong negative relationship between fairness concerns and willingness to pay for work meaning. Thus, individuals who prioritize fairness are less likely to accept lower wages for meaningful work. |
Keywords: | work meaning, labor supply, fairness preferences |
JEL: | C83 C90 M52 |
Date: | 2025 |
URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12068 |
By: | Khushboo Surana |
Abstract: | We propose a nonparametric method to identify the degree of heterogeneity in individual preferences. Using preference information revealed by observed behavior, the method estimates interpersonal preference heterogeneity as the Kemeny distance between individual preference rankings. Using data from the U.S. Panel Study of Income Dynamics (PSID), we derive bounds on the distance-based heterogeneity measure, which we then use to group individuals with similar preferences. We demonstrate that constructing such preference types can substantially strengthen empirical analysis by (i) producing more accurate demand predictions, (ii) improving welfare analysis, and (iii) detecting functional misspecifications in parametric methodologies. |
Keywords: | Preference heterogeneity, Revealed preference analysis, Kemeny distance |
JEL: | C14 C60 D11 D12 |
Date: | 2025–09 |
URL: | https://d.repec.org/n?u=RePEc:yor:yorken:25/03 |
By: | Atiyeh Yeganloo; Cahal Moran; Juvaria Jafri |
Keywords: | Charitable giving, donation, public goods, choice overload, choice deprivation, satisfaction, regret |
JEL: | C91 D64 D91 H00 |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:enp:wpaper:eprg2518 |
By: | Hendrik Rommeswinkel |
Abstract: | The paper characterizes the Shannon (1948) and Tsallis (1988) entropies in a standard framework of decision theory, mixture sets. Procedural mixture sets are introduced as a variant of mixture sets in which it is not necessarily true that a mixture of two identical elements yields the same element. This allows the process of mixing itself to have an intrinsic value. The paper proves the surprising result that simply imposing the standard axioms of von Neumann-Morgenstern on preferences on a procedural mixture set yields the entropy as a representation of procedural value. An application of the theorem to decision processes and the relation between choice probabilities and decision times elucidates the difficulty of extending the drift-diffusion model to multi-alternative choice. |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.07588 |
By: | Mohammad T. Hajiaghayi; Suho Shin |
Abstract: | The problem of delegated choice has been of long interest in economics and recently on computer science. We overview a list of papers on delegated choice problem, from classic works to recent papers with algorithmic perspectives. |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.06562 |
By: | Botelho Azevedo, Alda (Universidade Nova de Lisboa); Gonçalves, Inês (Universidade Nova de Lisboa); Pereira dos Santos, João (Queen Mary University of London) |
Abstract: | Our study investigates public opinion on the housing affordability crisis in Portugal through a nationally representative survey combined with an information provision experiment. Participants were asked to identify perceived causes of rising housing prices, assess their factual knowledge of the housing market and sociodemographic trends, and indicate their preferred policy solutions, carefully framed to reflect trade-offs. Half of the respondents were randomly assigned to receive official statistical information on these trends before indicating their policy preferences. The findings reveal significant heterogeneity in beliefs about the causes of the crisis, pervasive misperceptions regarding market trends, and a limited impact of information provision on policy preferences. These results underscore the challenges of addressing housing policy through informational interventions alone and highlight the need for strategies that integrate behavioral and contextual factors to foster informed public engagement. |
Keywords: | information provision experiment, housing, Portugal |
JEL: | R31 F60 H1 |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18073 |
By: | Richard S. J. Tol |
Abstract: | There are many published estimates of the social cost of carbon. Some are clear outliers, the result of poorly constrained models. Percentile winsorizing is an option, but I here propose conceptual winsorizing: The social cost of carbon is either a willingness to pay, which cannot exceed the ability to pay, or a proposed carbon tax, which cannot raise more revenue than all other taxes combined. Conceptual winsorizing successfully removes high outliers. It slackens as economies decarbonize, slowly without climate policy, faster with. |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.07384 |
By: | Hui-Kuan Chung; Nick Doren; Lasse Mononen; Mia Lu; Marcus Grueschow; Helen Hayward Könnecke; Alexander Jetter; Boris B. Quednow; Nick Netzer; Philippe N. Tobler |
Abstract: | Models of limited attention have the potential to become a new unifying paradigm that could replace the rational choice approach. In this paper, we test the limited attention hypothesis by enhancing attention using pharmacological substances. A total of 160 subjects participated in our randomized, placebo-controlled, and double-blind experimental study. We find that enhancing attention through boosting the noradrenergic system with reboxetine improves the quality of choice as captured by multiple different measures of rationality. Eye-tracking suggests that boosting noradrenaline promotes more rational choice by efficiently directing attention to more valuable options. Other attention-enhancing drugs (methylphenidate, which boosts the dopaminergic system, and nicotine, which boosts the cholinergic system) improve rationality to a lesser extent. Aside from testing the limited attention hypothesis directly, our results have implications for welfare economics, policy-design, and public health. |
Keywords: | limited attention, rationality, pharmacology |
JEL: | B41 C91 D01 D60 D91 |
Date: | 2025 |
URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12078 |
By: | Chikazoe, Junichi; Kawaguchi, Kohei; Suzuki, Kanji; Uetake, Kosuke; Watanabe, Yasutora; Yamada, Katsunori |
Abstract: | Default nudges are widely used and effective, but their mechanisms remain unclear. We test whether ease, endowment, or endorsement effects drive choices. In an online randomized experiment, the endowment channel emerges as the principal driver. We then use a novel fMRI approach that constructs brain activity maps of cognitions and uses them to trace their variation in each cognition during decision-making. This approach validates treatments by confirming they elicit the intended cognitions and uses them as instruments to identify the causal effect of cognition on choice. Results show that endowment drives default nudge effectiveness, suggesting policy designs should leverage it. |
Date: | 2025–09–01 |
URL: | https://d.repec.org/n?u=RePEc:osf:socarx:wfrsp_v1 |
By: | M. Pomazanov |
Abstract: | The key indicators of model stability are the population stability index (PSI), which uses the difference in population distribution, and the Kolmogorov-Smirnov statistic (KS) between two distributions. When deriving a binary choice model, the question arises about the real Gini index for any new model. The paper shows that when the Gini changes, the real Gini index should be less than the obtained Gini index. This type is included in the equation using a formula, and the PSI formula in KS is also included based on the scoring indicator. The error in calculating the Gini index of the equation is unavoidable, so it is necessary to always rely on the calculation formula. This type of research is suitable for a wide range of tasks where it is necessary to consider the error in scoring the indicator at any length. |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.04866 |
By: | Tal Gross; Timothy Layton; Daniel Prinz; Julia Yates |
Abstract: | We study how couples in the Medicare Part D program choose an insurance plan. Over 70 percent of enrollees choose the same plan as their spouse. Even among those with differing healthcare needs, well over half do so. Discrete-choice models suggest beneficiaries value being on the same plan as their spouse at over a thousand dollars per year. Using a regression-discontinuity design, we show that younger spouses disproportionately follow their older spouse’s plan choice. Joint plan choice contributes modestly to overall overspending, but increases costs substantially for the couples with different cost-minimizing plans. |
JEL: | D12 D14 I13 |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34137 |
By: | Karlsson, Niklas (Örebro University School of Business); Lunander, Anders (Örebro University School of Business) |
Abstract: | Given the presence of a cutoff score in a multiple-choice questions test, a challenge for the test maker is to choose a scoring method maximizing the probability of a passing score for those with adequate knowledge given a prescribed risk of passing those with insufficient understanding. Within the environment of a true-false choice test, we analyze the statistical power of the standard method - one point if the correct answer is marked and zero otherwise – with that of the negative marking method - no answer results in zero points, a correct answer generates one point, and an incorrect answer is penalized by one point. Our comparison of power between the two methods indicates that the power is about equal when test taker exhibits a small variance in terms of her degree of confidence across the questions. For larger variance, the negative marking method is superior to the standard method. However, the more the test taker fails to capture her level of confidence, i.e., mis-calibration of knowledge, the lower statistical power of the negative marking. Which method has the highest power depends on the magnitude of mis-calibration. Underrating does not affect the power of NM as much as overrating |
Keywords: | multiple-choice questions; negative marking; test of statistical power |
JEL: | A22 C12 |
Date: | 2025–08–28 |
URL: | https://d.repec.org/n?u=RePEc:hhs:oruesi:2025_009 |
By: | Shashwat Khare; Souvik Roy; Ton Storcken |
Abstract: | This paper studies matching markets where institutions are matched with possibly more than one individual. The matching market contains some couples who view the pair of jobs as complements. First, we show by means of an example that a stable matching may fail to exist even when both couples and institutions have responsive preferences. Next, we provide conditions on couples' preferences that are necessary and sufficient to ensure a stable matching for every preference profile where institutions may have any responsive preference. Finally, we do the same with respect to institutions' preferences, that is, we provide conditions on institutions' preferences that are necessary and sufficient to ensure a stable matching for every preference profile where couples may have any responsive preference. |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.07501 |
By: | Pedro Picchetti |
Abstract: | This paper studies the partial identification of treatment effects in Instrumental Variables (IV) settings with binary outcomes under violations of independence. I derive the identified sets for the treatment parameters of interest in the setting, as well as breakdown values for conclusions regarding the true treatment effects. I derive $\sqrt{N}$-consistent nonparametric estimators for the bounds of treatment effects and for breakdown values. These results can be used to assess the robustness of empirical conclusions obtained under the assumption that the instrument is independent from potential quantities, which is a pervasive concern in studies that use IV methods with observational data. In the empirical application, I show that the conclusions regarding the effects of family size on female unemployment using same-sex sibling as the instrument are highly sensitive to violations of independence. |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.10242 |
By: | Tingting Cheng; Jiachen Cong; Fei Liu; Xuanbin Yang |
Abstract: | In this paper, we propose a novel factor-augmented forecasting regression model with a binary response variable. We develop a maximum likelihood estimation method for the regression parameters and establish the asymptotic properties of the resulting estimators. Monte Carlo simulation results show that the proposed estimation method performs very well in finite samples. Finally, we demonstrate the usefulness of the proposed model through an application to U.S. recession forecasting. The proposed model consistently outperforms conventional Probit regression across both in-sample and out-of-sample exercises, by effectively utilizing high-dimensional information through latent factors. |
Date: | 2025–07 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.16462 |
By: | Toger, Marina; Türk, Umut; Östh, John; Fischer, Manfred M. |
Abstract: | This study applies a heteroscedastic spatial Durbin panel data model to investigate how sociodemographic and socioeconomic factors influence regional commuter mobility in the Greater Stockholm Area. Commuter mobility, defined as the flow of people to and from workplaces across regions and over time, is measured using high-frequency, high-resolution origin-destination data derived from mobile phone records, providing high-frequency, high-resolution insights into commuting patterns. The analysis uses a balanced panel of 681 regions from 2018–2024, incorporating an 18-nearest-neighbor spatial weight matrix to capture the topological relationships. Direct (withinregion) and indirect (spillover) effects are estimated using Bayesian inference, enabling robust interpretation of marginal effects in the presence of spatial lags in dependent and independent variables. Results show that spatial spillovers exert a more decisive influence than direct effects, with educational attainment and car ownership emerging as the most influential determinants of commuter mobility. According to total impact estimates, the demographic structure plays a comparatively minor, yet still significant, role. |
Keywords: | Spatial econometrics; Bayesian estimation; heteroscedastic spatial Durbin panel data model; GSM- based mobility flow data, ; spatial spillover effects; Greater Stockholm Area |
Date: | 2025–08–22 |
URL: | https://d.repec.org/n?u=RePEc:wiw:wus046:76951486 |
By: | Bonacina, Monica; Demir, Mert; Sileo, Antonio; Zanoni, Angela |
Abstract: | The transition to a zero-emission vehicle fleet represents a pivotal element of Europe’s decarbonization strategy, with Italy’s participation being particularly significant given the size of its automotive market. This study investigates the potential for battery electric cars (BEVs) to drive decarbonization of Italy’s passenger vehicle fleet, focusing on the feasibility of targets set in the National Integrated Plan for Energy and Climate (PNIEC). Leveraging artificial neural networks, we integrate macroeconomic indicators, market-specific variables, and policy instruments to predict fleet dynamics and identify key factors influencing BEV adoption. We forecast that while BEV registrations will continue growing through 2030, the growth rate is projected to decelerate, presenting challenges for meeting ambitious policy targets. Our feature importance analysis demonstrates that BEV adoption is driven by an interconnected set of economic, infrastructural, and behavioral factors. Specifically, our model highlights that hybrid vehicle registrations and the vehicle purchase index exert the strongest influence on BEV registrations, suggesting that policy interventions should prioritize these areas to maximize impact. By offering data-driven insights and methodological innovations, our findings contribute to more effective policy design for accelerating sustainable mobility adoption while accounting for market realities and consumer behavior. |
Keywords: | Climate Change, Environmental Economics and Policy, Sustainability |
Date: | 2025–08–01 |
URL: | https://d.repec.org/n?u=RePEc:ags:feemwp:369002 |
By: | Robin Ng; Greg Taylor |
Abstract: | We study how content moderation facilitates communication on online platforms. A sender transmits information to a receiver, exerting effort to signal their truthfulness. Communication fails without moderation because the effort required is prohibitive. Moderation resolves this problem by making effort a more powerful signal of veracity. However, moderation crowds-out sender effort, decreasing content quality on the platform. A socially optimal or profit-maximizing policy may therefore involve limited moderation. We study the choice between being a platform or broadcaster, how moderation influences competition for attention, and the effects of misinformation actors, AI-generated content, and moderator errors on the sustainability of communication. |
Keywords: | user-generated content, content moderation, creator economy, media platforms, misinformation |
JEL: | D83 L82 L86 |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:bon:boncrc:crctr224_2025_698 |
By: | Kiuchi, Keita (National Institute of Occupational Safety and Health, Japan) |
Abstract: | Multimodal emotion recognition, integrating facial expressions and vocal features, is key to advancing human-computer interaction and mental healthcare. However, current deep learning models often lack interpretability, limiting their real-world applicability. In this study, we present a framework for emotion recognition (happy, sad, angry, neutral) that uncovers meaningful multimodal patterns. We collected online dialogue data from 99 participants, extracting facial features (OpenFace), acoustic descriptors (openSMILE), and deep audio embeddings (VGGish). Factor analysis (FA) was applied independently to each modality for dimensionality reduction, and Gaussian mixture modeling (GMM) on the combined factor scores revealed latent multimodal expression clusters. These cluster probabilities, along with participant covariates (e.g., Big Five traits; PHQ-9; GAD-7/SCAS), served as inputs to various classifiers, including XGBoost and random forest. SHAP analysis confirmed the interpretability of the clusters, illustrating how individual differences and covariates influenced emotion predictions. We identified 15 distinct expression clusters (e.g., “social smiles, ” “inexpressive states, ” “wry grins”), offering nuanced insights into affective displays. Although overall accuracy was modest—due to individual variability and label noise—the framework effectively highlights interpretable, fine-grained expressive patterns. This approach lays groundwork for transparent affective computing systems, such as empathetic conversational agents, emphasizing the importance of explainability in emotion-based applications. |
Date: | 2025–08–24 |
URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:2udp8_v1 |
By: | Robin Musolff; Florian Zimmermann |
Abstract: | Mental models help people navigate complex environments. This paper studies how people deal with model uncertainty. In an experiment, participants estimate a company’s value, facing uncertainty about which one of two models correctly determines its true value. Using a between subjects design, we vary the degree of model complexity. Results show that in high-complexity conditions people fully neglect model uncertainty in their actions. However, their beliefs continue to reflect model uncertainty. This disconnect between beliefs and actions suggests that complexity leads to biased decision-making, while beliefs remain more nuanced. Furthermore, we show that complexity, via full uncertainty neglect, leads to higher confidence in the optimality of own actions. |
Keywords: | Mental Models, Beliefs, Attention, Confidence, Representations |
JEL: | D01 D83 |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:bon:boncrc:crctr224_2025_697 |
By: | Hendrik Rommeswinkel |
Abstract: | Decision makers may face situations in which they cannot observe the consequences that result from their actions. In such decisions, motivations other than the expected utility of consequences may play a role. The present paper axiomatically characterizes a decision model in which the decision maker cares about whether it can be ex post verified that a good consequence has been achieved. Preferences over acts uniquely characterize a set of events that the decision maker expects to be able to verify in case they occur. The decision maker chooses the act that maximizes the expected utility across verifiable events of the worst possible consequence that may have occurred. For example, a firm choosing between different carbon emission reduction technologies may find some technologies to leave ex post more uncertainty about the level of emission reduction than other technologies. The firm may care about proving to its stakeholders that a certain amount of carbon reduction has been achieved and may employ privately obtained evidence to do so. It may choose in expectation less efficient technologies if the achieved carbon reduction is better verifiable using the expected future evidence. |
Date: | 2025–08 |
URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.19585 |