|
on Discrete Choice Models |
| By: | Bas Donkers; Kamel Jedidi; Miłosz Kadziński; Mohammad Ghaderi |
| Abstract: | We introduce the Random Preference Model (RPM), a non-parametric and flexible discrete choice model. RPM is a rank-based stochastic choice model where choice options have multi-attribute representations. It takes preference orderings as the main primitive and models choices directly based on a distribution over partial or complete preference orderings over a finite set of alternatives. This enables it to capture context-dependent behaviors while maintaining adherence to the regularity axiom. In its output, it provides a full distribution over the entire preference parameter space, accounting for inferential uncertainty due to limited data. Each ranking is associated with a subspace of utility functions and assigned a probability mass based on the expected log-likelihood of those functions in explaining the observed choices. We propose a two-stage estimation method that separates the estimation of ranking-level probabilities from the inference of preference parameters variation for a given ranking, employing Monte Carlo integration with subspace-based sampling. To address the factorial complexity of the ranking space, we introduce scalable approximation strategies: restricting the support of RPM to a randomly sampled or orthogonal basis subset of rankings and using partial permutations (top-k lists). We demonstrate that RPM can effectively recover underlying preferences, even in the presence of data inconsistencies. The experimental evaluation based on real data confirms RPM variants consistently outperform multinomial logit (MNL) in both in-sample fit and holdout predictions across different training sizes, with support-restricted and basis-based variants achieving the best results under data scarcity. Overall, our findings demonstrate RPM's flexibility, robustness, and practical relevance for both predictive and explanatory modeling. |
| Keywords: | choice models, context-dependent preference, nonparametric modeling, random utility, rankings |
| JEL: | C35 C14 C15 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:bge:wpaper:1502 |
| By: | Reich, Charlotte; Bae, Dongjin; Mußhoff, Oliver; Bruns, Selina J. K. |
| Abstract: | Transformation of the food system is critical in the face of growing challenges such as climate change. Smallholder farmers in the Global South are particularly vulnerable to these challenges, often living in poverty. One potential pathway to increasing their resilience and income is through price premiums and consistent buy-offs from supermarkets. However, supermarkets rarely source from domestic smallholder farmers due to unstable production and a lack of guarantee of complying with food safety standards. An instrument to overcome these shortcomings are certification schemes. However, for a farmer to invest in or for a policy to subsidize certification, it is central to understand if there is a consumer group that will respond to it. Thus, our objective is to investigate whether urban consumers in a low-income country setting are willing to pay a premium for certified food. We specifically focus on Cambodia and the newly established Cambodian Good Agricultural Practice (CamGAP) certification, which promotes food safety. This research seeks to understand whether the willingness to pay (WTP) of consumers can support the entry of small farmers into the supermarket supply chain. In markets where food safety is an emerging concern but clear signals are lacking, a key question is whether certification can override existing heuristics consumers currently use to identify safe food. We used a quantitative questionnaire and a discrete choice experiment (DCE) with middle to high income urban shoppers to assess their WTP for CamGAP certified food. We also conducted an information treatment as part of the DCE by showing a short video to a randomly selected treatment group to examine the impact of consumer knowledge on purchasing decisions. Our results show that consumers' WTP is higher for certified food, with an even stronger WTP after receiving the information treatment. Notably, consumers were also more willing to pay for domestic fresh food after the treatment. Our study provides first evidence for governments and international agencies that certification can support smallholder farmers in accessing supermarkets, even in countries where food safety certifications are just entering the market. |
| Keywords: | Smallholder Farmers, Discrete Choice Experiment, Willingness to Pay, Market Integration |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:daredp:323233 |
| By: | Héctor Juan Villarreal Páez (School of Government and Public Transformation, Tecnológico de Monterrey); Diego Vázquez-Pimentel (School of Government and Public Transformation, Tecnológico de Monterrey) |
| Abstract: | This paper explores the dynamics of labor informality in Mexico by developing a discrete choice mixed logit model to explain the transitions between labor states—namely, not employed, formal employment, and informal employment—among individuals aged 18 to 65. The study offers critical insights into the informal sector’s heterogeneity, with particular focus on voluntary versus involuntary informality, while also contributing a novel estimation strategy that combines supply- and demand-side constraints within the informal labor market. The results highlight the persistent barriers to formal employment for a significant segment of the labor force, despite policy efforts aimed at reducing informality in Mexico. |
| Keywords: | informality, restricted choices, microsimulations, labor policies |
| JEL: | J24 J31 C54 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:gnt:wpaper:5 |
| By: | Blackman, Allen; Jeuland, Marc; Leguizamo, Emilio |
| Abstract: | The ability of countries in Latin America to achieve net zero greenhouse gas (GHG) emissions by mid-century, the target set by the Paris Agreement, will depend critically on citizen support. To gauge this support, we administered a contingent valuation survey to representative samples in seven of the regions leading GHG emitting countries and in the United States, which is used as a comparator. The survey elicits respondents willingness to pay (WTP) for achieving net zero by 2050 and uses a split sample design to test whether WTP is affected by the distribution of decarbonization costs across households. Our estimates of mean WTP in the Latin American study countries are on par both with our estimate for the United States, and with estimates from a recent CV study for China, Sweden, and the United States. However, among the Latin American study countries, mean WTPs for Argentina and Brazil are relatively low. We also find that the distribution of the costs of decarbonization across households does not have a clear effect on WTP and that the drivers of WTP for our Latin American study countries are similar to those the literature has identified in other regions. |
| Keywords: | Contingent Valuation;stated preference;Net Zero;Argentina;Brazil;Chile;Colombia;Ecuador;Mexico;Peru |
| JEL: | Q51 Q54 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:idb:brikps:14188 |
| By: | Shyamal Chowdhury; Manuela Puente-Beccar; Hannah Schildberg-Hörisch; Sebastian O. Schneider; Matthias Sutter |
| Abstract: | We investigate how strongly the local environment beyond the family can contribute to understanding the formation of children's economic preferences. Building on precise geolocation data for around 6.000 children, we use fixed effects, spatial autoregressive models and Kriging to capture the relation between the local environment and children's preferences. The spatial models explain a considerable part of so far unexplained variation in preferences. Moreover, the "spatial stability" of preferences exceeds the village level. Our results highlight the importance of the local environment for the formation of children's preferences, which we quantify to be as large as that of parental preferences. |
| Keywords: | skill formation, spatial models, kriging, local environment, patience, risk attitudes, prosociality, experiments with children, Bangladesh |
| JEL: | D01 C21 C99 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12001 |
| By: | Joshua Foster; Fredrik Odegaard |
| Abstract: | This paper proposes a new demand estimation method using attention-based language models. An encoder-only language model is trained in a two-stage process to analyze the natural language descriptions of used cars from a large US-based online auction marketplace. The approach enables semi-nonparametrically estimation for the demand primitives of a structural model representing the private valuations and market size for each vehicle listing. In the first stage, the language model is fine-tuned to encode the target auction outcomes using the natural language vehicle descriptions. In the second stage, the trained language model's encodings are projected into the parameter space of the structural model. The model's capability to conduct counterfactual analyses within the trained market space is validated using a subsample of withheld auction data, which includes a set of unique "zero shot" instances. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.17564 |
| By: | Eric Innocenti (LISA - Laboratoire « Lieux, Identités, eSpaces, Activités » (UMR CNRS 6240 LISA) - CNRS - Centre National de la Recherche Scientifique - Università di Corsica Pasquale Paoli [Université de Corse Pascal Paoli]); Dominique Prunetti (LISA - Laboratoire « Lieux, Identités, eSpaces, Activités » (UMR CNRS 6240 LISA) - CNRS - Centre National de la Recherche Scientifique - Università di Corsica Pasquale Paoli [Université de Corse Pascal Paoli]); Marielle Delhom (SPE - Laboratoire « Sciences pour l’Environnement » (UMR CNRS 6134 SPE) - CNRS - Centre National de la Recherche Scientifique - Università di Corsica Pasquale Paoli [Université de Corse Pascal Paoli]); Corinne Idda (LISA - Laboratoire « Lieux, Identités, eSpaces, Activités » (UMR CNRS 6240 LISA) - CNRS - Centre National de la Recherche Scientifique - Università di Corsica Pasquale Paoli [Université de Corse Pascal Paoli]) |
| Abstract: | Agent-Based Models focusing on land markets provide a computational framework to simulate socio-economic dynamics in land and real-estate markets. In this paper, we introduce the 5-Step Simulation Iterative Modelling Process method, an iterative, five-step modelling and simulation decomposition approach specifically designed to structure the development of Agent-Based Land Market Models. We describe how implementing reusable building blocks-conceptual, computational, and executable-enhances modularity and fosters reusability of both theoretical concepts and software code. An illustrative example, applied to land and real-estate markets in Corsica, concretely demonstrates the application of the method and the creation of these reusable components. The integration of the economic concepts of Willingness To Accept and Willingness To Pay into the design of an Agent-Based Land Market Model exemplifies how these building blocks contribute to market dynamics formation. Finally, we highlight the potential of this approach to strengthen computational simulation, support socio-economic analysis, and promote sustainable land management. |
| Keywords: | Agent-Based Modelling, Land and Real Estate Markets, Reusable Building Blocks, Socio-Economic Dynamics, Sustainable Land Management |
| Date: | 2025–06–30 |
| URL: | https://d.repec.org/n?u=RePEc:hal:journl:hal-05161281 |
| By: | Wang, Peggy PhD |
| Abstract: | Researchers at UC Berkeley conducted semi-structured interviews with 15 visually impaired individuals. They exploredtheir perspectives regarding current travel behavior and transportation experience, and the potential of Shared Automated Vehicles (SAVs) to enhance their travel experiences and address existing transportation challenges. The results revealed a range of expectations and concerns related to SAVs, particularly in the areas of accessibility, safety, communication, and affordability. Most participants expressed enthusiasm for the potential benefits of SAVs to increase independence and access to underserved areas. They also highlighted critical accessibility needs, such as reliable vehicle identification, accurate drop-off locations, clear communication channels, and accessible interfaces. Affordability emerged as a key factor influencing potential SAV adoption, with many participants indicating a preference for SAVs if they were priced competitively with existing transportation options, especially rideshare services. The findings of this study provide valuable insights for policymakers, transportation planners, and SAV developers to ensure that future autonomous transportation solutions are truly inclusive and meet the diverse needs of visually impaired travelers. |
| Keywords: | Engineering, Accessibility, Shared Automated Vehicles, Visually Impaired Travelers |
| Date: | 2025–07–01 |
| URL: | https://d.repec.org/n?u=RePEc:cdl:itsrrp:qt58w5v9x1 |
| By: | Chiara Farronato; Andrey Fradkin; Tesary Lin |
| Abstract: | We study the welfare consequences of choice architecture for online privacy using a field experiment that randomizes cookie consent banners. We study three ways in which firms or policymakers can influence choices: (1) nudging users through banner design to encourage acceptance of cookie tracking; (2) setting defaults when users dismiss banners; and (3) implementing consent decisions at the website versus browser level. Absent design manipulation, users accept all cookies more than half of the time. Placing cookie options behind extra clicks strongly influences choices, shifting users toward more easily accessible alternatives. Many users dismiss banners without making an explicit choice, underscoring the importance of default settings. Survey evidence further reveals substantial confusion about default settings. Using a structural model, we find that among consent policies requiring site-specific decisions, consumer surplus is maximized when consent interfaces clearly display all options and default to acceptance in the absence of an explicit choice. However, the welfare gains from optimizing banner design are much smaller than those from adopting browser-level consent, which eliminates the time costs of repeated decisions. |
| JEL: | C93 D18 D83 D91 K20 L51 L86 |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34025 |
| By: | McKenzie, David |
| Abstract: | This paper offers practical advice on how to improve statistical power in randomized experiments through choices and actions researchers can take at the design, implementation, and analysis stages. At the design stage, the choice of estimand, choice of treatment, and decisions that affect the residual variance and intra-cluster correlation can all affect power for a given sample size. At the implementation stage, researchers can boost power through increasing compliance with treatment, reducing attrition, and improving outcome measurement. At the analysis stage, power can be increased through using different test statistics or estimands, through the choice of control variables, and through incorporating informative priors in a Bayesian analysis. A key message is that it does not make sense to talk of “the” power of an experiment. A study can be well-powered for one outcome or estimand, but not others, and a fixed sample size can yield very different levels of power depending on researcher decisions. |
| Date: | 2025–07–21 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wbrwps:11176 |
| By: | Martin Obradovits; Markus Walzl |
| Abstract: | Consumers increasingly value the environmental and social responsibility of the production processes used by firms, yet these processes often remain unobservable, even after consumption. In this paper, we develop a simple model to examine firms’ technology choices and subsequent price competition in markets for such label credence goods with hidden process attributes. Using a multi-sender signaling framework, we show that in the payoff-dominant equilibrium, firms can partially signal their production choices and avoid Bertrand competition when at least one firm adopts a green technology. Surprisingly, increasing consumers’ environmental concern or eliminating the information asymmetry may reduce social welfare by discouraging green production. |
| Keywords: | label credence goods, technology choice, asymmetric information, price competition, signaling, green production |
| JEL: | D82 D83 L13 L15 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:jku:econwp:2025-11 |
| By: | Haochen Luo; Yuan Zhang; Chen Liu |
| Abstract: | Sparse portfolio optimization is a fundamental yet challenging problem in quantitative finance, since traditional approaches heavily relying on historical return statistics and static objectives can hardly adapt to dynamic market regimes. To address this issue, we propose Evolutionary Factor Search (EFS), a novel framework that leverages large language models (LLMs) to automate the generation and evolution of alpha factors for sparse portfolio construction. By reformulating the asset selection problem as a top-m ranking task guided by LLM-generated factors, EFS incorporates an evolutionary feedback loop to iteratively refine the factor pool based on performance. Extensive experiments on five Fama-French benchmark datasets and three real-market datasets (US50, HSI45 and CSI300) demonstrate that EFS significantly outperforms both statistical-based and optimization-based baselines, especially in larger asset universes and volatile conditions. Comprehensive ablation studies validate the importance of prompt composition, factor diversity, and LLM backend choice. Our results highlight the promise of language-guided evolution as a robust and interpretable paradigm for portfolio optimization under structural constraints. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.17211 |
| By: | Xu Cheng (University of Pennsylvania); Frank Schorfheide (University of Pennsylvania); Peng Shao (Auburn University) |
| Abstract: | This paper studies the estimation of multi-dimensional heterogeneous parameters in a nonlinear panel data model with endogeneity. These heterogeneous parameters are modeled with group patterns. Through estimating multiple memberships for each unit, the proposed method is robust to limited information from a subset of clusters: either due to sparse interactions of characteristics or weak identification of some combinations of heterogeneous parameters. We estimate the memberships along with the group specific and common parameters in a nonlinear GMM framework and derive their large sample properties. Finally, we apply this approach to the estimation of heterogeneous firm-level production functions parameters which are converted into markup estimates. |
| Keywords: | Clustering, GMM, K-means, Panel Data, Production Function Estimation |
| JEL: | C13 C23 D22 D24 E23 |
| Date: | 2025–06–19 |
| URL: | https://d.repec.org/n?u=RePEc:pen:papers:25-014 |
| By: | Holt, Stephen B.; Vinopal, Katie |
| Abstract: | Holt and Vinopal (2023) provides evidence of an income-based gap in time spent waiting for services on the typical day in the United States. The gap was estimated at the extensive margin, unconditional intensive margin, and intensive margin conditional on some waiting time. The analysis was complemented with a heterogeneity assessment that found the income-based gap in the length of waiting spells conditional on some waiting was observed across race categories except Black Americans. Unfortunately, Hall and Thiele identified an error in the code regarding missing household income data that led to a larger analytic sample than was warranted. In addition, Hall and Thiele claim some sensitivity in the hypothesis tests of some results to an expanded definition of the dependent variable and standard error estimation choice. However, correcting our original coding errors regarding missing income, adding weights when omitted, and adding controls when omitted does not meaningfully alter the estimated coefficients or their precision. Hall and Thiele's claims that our race results are not robust is misleading, as they seem driven by interpretation of simplified and space-constrained writing in the abstract, rather than lack of reproduction or sensitivity of the empirical results. To the extent we observe some sensitivity, it involves Hall and Thiele's move to an expansive definition of waiting time to include more leisure and entertainment services, a change that substantially alters the interpretation of results and goes beyond a simple test of robustness. |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:i4rdps:244 |