|
on Discrete Choice Models |
| By: | Abate, Gashaw T.; Bernard, Tanguy; Deutschmann, Joshua; Fall, Fatou |
| Abstract: | Individuals often make decisions considering both private returns and welfare impacts on others. Food safety decisions by smallholder agricultural producers exemplify this choice, particularly in low-income countries where farmers often consume some of the food crops they produce and sell or donate the rest. We conduct a lab-in-the-field experiment with peanuts producers in Senegal to study the decision to invest in food safety information, exogenously varying the degree of private returns (monetary or health-wise) and welfare impacts on others. Producers are willing to pay real money for food safety information even absent the potential for private returns, but willingness to pay increases with the potential for private returns. A randomized information treatment significantly increases willingness to pay in all scenarios. Our results shed light on the complex interplay between altruism and economic decisions in the presence of externalities, and point to the potential of timely and targeted information to address food safety issues. |
| Keywords: | food safety; health; groundnuts; aflatoxins; smallholders; returns; Senegal; Africa; Sub-Saharan Africa; Western Africa |
| Date: | 2025–07–07 |
| URL: | https://d.repec.org/n?u=RePEc:fpr:ifprid:175569 |
| By: | Chris Engh |
| Abstract: | We study a generalization of the state-action rational inattention model with two dimensions of uncertainty: states and hedonic characteristics. The resulting conditional choice probability takes a familiar weighted multinomial logit form, but in contrast to the state-action model, is unique and is not a corner solution. The model imposes testable restrictions on the behavior of choice probabilities in markets with product entry. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.05939 |
| By: | Georges Sfeir; Gabriel Nova; Stephane Hess; Sander van Cranenburgh |
| Abstract: | Large Language Models (LLMs) are widely used to support various workflows across different disciplines, yet their potential in choice modelling remains relatively unexplored. This work examines the potential of LLMs as assistive agents in the specification and, where technically feasible, estimation of Multinomial Logit models. We implement a systematic experimental framework involving thirteen versions of six leading LLMs (ChatGPT, Claude, DeepSeek, Gemini, Gemma, and Llama) evaluated under five experimental configurations. These configurations vary along three dimensions: modelling goal (suggesting vs. suggesting and estimating MNLs); prompting strategy (Zero-Shot vs. Chain-of-Thoughts); and information availability (full dataset vs. data dictionary only). Each LLM-suggested specification is implemented, estimated, and evaluated based on goodness-of-fit metrics, behavioural plausibility, and model complexity. Findings reveal that proprietary LLMs can generate valid and behaviourally sound utility specifications, particularly when guided by structured prompts. Open-weight models such as Llama and Gemma struggled to produce meaningful specifications. Claude 4 Sonnet consistently produced the best-fitting and most complex models, while GPT models suggested models with robust and stable modelling outcomes. Some LLMs performed better when provided with just data dictionary, suggesting that limiting raw data access may enhance internal reasoning capabilities. Among all LLMs, GPT o3 was uniquely capable of correctly estimating its own specifications by executing self-generated code. Overall, the results demonstrate both the promise and current limitations of LLMs as assistive agents in choice modelling, not only for model specification but also for supporting modelling decision and estimation, and provide practical guidance for integrating these tools into choice modellers' workflows. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.21790 |
| By: | Inoue, Chihiro; Saito, Asumi; Takahashi, Yuki (Tilburg University, Center For Economic Research) |
| Keywords: | STEM Gender Gap; college choice; gender ratio; preference elicitation; discrete choice experiment |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:tiu:tiucen:d9c3a116-f15b-48b7-b86d-906a7a95f938 |
| By: | Kretz, Claudio; Puppe, Clemens |
| Abstract: | We reappraise the Arrow problem by studying the aggregation of choice functions. We do so in the general framework of judgment aggregation, in which choice functions are naturally representable by specifying, for each menu A and each alternative x in A, whether x is choosable from A, or not. Our framework suggests a natural strengthening of Arrow's independence condition positing that the collective choosability of an alternative from a menu depends on the individual views on that issue, and that issue alone. Our analysis reveals that Arrovian impossibility results crucially hinge on what internal consistency requirements we impose on choice functions. While the aggregation of 'binary' choice functions, i.e. those satisfying both contraction (») and expansion (Ú) consistency, is necessarily dictatorial, possibilities in the form of oligarchic rules emerge for path-independent choice functions, that is, when the expansion property Ú is replaced by the so-called Aizerman condition. Remarkably, the Arrovian aggregation of choice functions is shown to be almost dictatorial already under property Ú alone. When giving up expansion consistency, specific quota rules become possible. |
| Keywords: | choice function, rationalizability, aggregation theory, independence, Arrow's Theorem |
| JEL: | D01 D71 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:kitwps:323218 |
| By: | Elisabeth Beckmann; Söhnke Bergmann; Christa Hainz; Sarah Kiesl-Reiter |
| Abstract: | Loan guarantees can enhance access to credit, but serving as a private guarantor may also increase financial vulnerability. We examine, through a randomized information experiment in the UK, how providing information about the legal ramifications and risks of loan guarantees affects individuals’ willingness to act as guarantors. We find that providing information about legal risks reduces the willingness to guarantee loans, with stronger effects for larger loan amounts. Social preferences influence individuals’ willingness to act as guarantors. Information about legal ramifications increases the willingness to grant a guarantee among altruists but decreases it among those high in positive reciprocity. While information about the UK default rate reduces willingness, individuals are less likely to update their expectations for someone they know personally, indicating in-group bias. |
| Keywords: | third-party loan guarantees, survey experiment, social preferences, loan default expectations |
| JEL: | D14 G41 G51 G53 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12022 |
| By: | Yingnan Yan; Tianming Liu; Yafeng Yin |
| Abstract: | As a key advancement in artificial intelligence, large language models (LLMs) are set to transform transportation systems. While LLMs offer the potential to simulate human travelers in future mixed-autonomy transportation systems, their behavioral fidelity in complex scenarios remains largely unconfirmed by existing research. This study addresses this gap by conducting a comprehensive analysis of the value of travel time (VOT) of a popular LLM, GPT-4o. We employ a full factorial experimental design to systematically examine the LLM's sensitivity to various transportation contexts, including the choice setting, travel purpose, income, and socio-demographic factors. Our results reveal a high degree of behavioral similarity between the LLM and humans. The LLM exhibits an aggregate VOT similar to that of humans, and demonstrates human-like sensitivity to travel purpose, income, and the time-cost trade-off ratios of the alternatives. Furthermore, the behavioral patterns of LLM are remarkably consistent across varied contexts. However, we also find that the LLM's context sensitivity is less pronounced than that observed in humans. Overall, this study provides a foundational benchmark for the future development of LLMs as proxies for human travelers, demonstrating their value and robustness while highlighting that their blunted contextual sensitivity requires careful consideration. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.22244 |
| By: | Somdeb Lahiri |
| Abstract: | We provide an axiomatic characterization of lexicographic preferences over the set of all random availability functions using two assumptions. The first assumption is strong monotonicity, which in our framework is equivalent to the strong dominance property in microeconomics. The second assumption is independence of worse alternatives and we show that a weaker version of the same suffices for our purpose. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.12997 |
| By: | Hoyoung Lee; Junhyuk Seo; Suhwan Park; Junhyeong Lee; Wonbin Ahn; Chanyeol Choi; Alejandro Lopez-Lira; Yongjae Lee |
| Abstract: | In finance, Large Language Models (LLMs) face frequent knowledge conflicts due to discrepancies between pre-trained parametric knowledge and real-time market data. These conflicts become particularly problematic when LLMs are deployed in real-world investment services, where misalignment between a model's embedded preferences and those of the financial institution can lead to unreliable recommendations. Yet little research has examined what investment views LLMs actually hold. We propose an experimental framework to investigate such conflicts, offering the first quantitative analysis of confirmation bias in LLM-based investment analysis. Using hypothetical scenarios with balanced and imbalanced arguments, we extract models' latent preferences and measure their persistence. Focusing on sector, size, and momentum, our analysis reveals distinct, model-specific tendencies. In particular, we observe a consistent preference for large-cap stocks and contrarian strategies across most models. These preferences often harden into confirmation bias, with models clinging to initial judgments despite counter-evidence. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.20957 |
| By: | Joel L. Horowitz; Sokbae Lee |
| Abstract: | This paper presents a computationally efficient method for binary classification using Manski's (1975, 1985) maximum score model when covariates are discretely distributed and parameters are partially but not point identified. We establish conditions under which it is minimax optimal to allow for either non-classification or random classification and derive finite-sample and asymptotic lower bounds on the probability of correct classification. We also describe an extension of our method to continuous covariates. Our approach avoids the computational difficulty of maximum score estimation by reformulating the problem as two linear programs. Compared to parametric and nonparametric methods, our method balances extrapolation ability with minimal distributional assumptions. Monte Carlo simulations and empirical applications demonstrate its effectiveness and practical relevance. |
| Date: | 2025–08–05 |
| URL: | https://d.repec.org/n?u=RePEc:azt:cemmap:16/25 |
| By: | Zeqi Wu; Meilin Wang; Wei Huang; Zheng Zhang |
| Abstract: | Estimation and inference of treatment effects under unconfounded treatment assignments often suffer from bias and the `curse of dimensionality' due to the nonparametric estimation of nuisance parameters for high-dimensional confounders. Although debiased state-of-the-art methods have been proposed for binary treatments under particular treatment models, they can be unstable for small sample sizes. Moreover, directly extending them to general treatment models can lead to computational complexity. We propose a balanced neural networks weighting method for general treatment models, which leverages deep neural networks to alleviate the curse of dimensionality while retaining optimal covariate balance through calibration, thereby achieving debiased and robust estimation. Our method accommodates a wide range of treatment models, including average, quantile, distributional, and asymmetric least squares treatment effects, for discrete, continuous, and mixed treatments. Under regularity conditions, we show that our estimator achieves rate double robustness and $\sqrt{N}$-asymptotic normality, and its asymptotic variance achieves the semiparametric efficiency bound. We further develop a statistical inference procedure based on weighted bootstrap, which avoids estimating the efficient influence/score functions. Simulation results reveal that the proposed method consistently outperforms existing alternatives, especially when the sample size is small. Applications to the 401(k) dataset and the Mother's Significant Features dataset further illustrate the practical value of the method for estimating both average and quantile treatment effects under binary and continuous treatments, respectively. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.04044 |
| By: | Tal Gross (Boston University); Tim Layton (University of Virginia); Dániel Prinz (Institute for Fiscal Studies); Julia Yates (University of Michigan) |
| Date: | 2025–08–07 |
| URL: | https://d.repec.org/n?u=RePEc:ifs:ifsewp:25/28 |
| By: | Zhi Hao Lim |
| Abstract: | Peer information is pervasive in the workplace, but workers differ in whether and why they value such information. We develop a portable, theory-driven methodology to study heterogeneity in information preferences and the underlying mechanisms. In a real-effort experiment with 793 workers, we elicit willingness-to-pay for peer information delivered either before or after a task. We identify four worker types (indifferent, stress-avoidant, competitive, and learning-oriented) whose effort responses align with theoretical predictions. Workers' stated motivations in free-text responses strongly correlate with their revealed preferences and behavior, validating our classification. Notably, a nontrivial share (15%) strictly prefers to avoid information ex ante due to stress and exhibit no productivity gains from it. Tailoring the timing of information by worker type improves welfare by up to 48% relative to a uniform policy. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.06162 |
| By: | Joel L. Horowitz (Institute for Fiscal Studies); Sokbae Lee (Institute for Fiscal Studies) |
| Date: | 2025–08–05 |
| URL: | https://d.repec.org/n?u=RePEc:ifs:ifsewp:cwp16/24 |
| By: | Amine Allouah; Omar Besbes; Josu\'e D Figueroa; Yash Kanoria; Akshit Kumar |
| Abstract: | Online marketplaces will be transformed by autonomous AI agents acting on behalf of consumers. Rather than humans browsing and clicking, vision-language-model (VLM) agents can parse webpages, evaluate products, and transact. This raises a fundamental question: what do AI agents buy, and why? We develop ACES, a sandbox environment that pairs a platform-agnostic VLM agent with a fully programmable mock marketplace to study this question. We first conduct basic rationality checks in the context of simple tasks, and then, by randomizing product positions, prices, ratings, reviews, sponsored tags, and platform endorsements, we obtain causal estimates of how frontier VLMs actually shop. Models show strong but heterogeneous position effects: all favor the top row, yet different models prefer different columns, undermining the assumption of a universal "top" rank. They penalize sponsored tags and reward endorsements. Sensitivities to price, ratings, and reviews are directionally human-like but vary sharply in magnitude across models. Motivated by scenarios where sellers use AI agents to optimize product listings, we show that a seller-side agent that makes minor tweaks to product descriptions, targeting AI buyer preferences, can deliver substantial market-share gains if AI-mediated shopping dominates. We also find that modal product choices can differ across models and, in some cases, demand may concentrate on a few select products, raising competition questions. Together, our results illuminate how AI agents may behave in e-commerce settings and surface concrete seller strategy, platform design, and regulatory questions in an AI-mediated ecosystem. |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2508.02630 |
| By: | Sinan Aral; Seth G Benzell; Avinash Collis; Christos Nicolaides |
| Abstract: | We use representative, incentive-compatible online choice experiments involving 19, 923 Facebook, Instagram, LinkedIn, and X users in the US to provide the first large-scale, empirical measurement of local network effects in the digital economy. Our analysis reveals social media platform value ranges from $78 to $101 per consumer, per month, on average, and that 20-34% of that value is explained by local network effects. We also find 1) stronger ties are more valuable on Facebook and Instagram, while weaker ties are more valuable on LinkedIn and X; 2) connections known through work are most valuable on LinkedIn and least valuable on Facebook, and people looking for work value LinkedIn significantly more and Facebook significantly less than people not looking for work; 3) men value connections to women on social media significantly more than they value connections to other men, particularly on Instagram, Facebook and X, while women value connections to men and women equally; 4) white consumers value relationships with other white consumers significantly more than they value relationships with non-white consumers on Facebook while, on Instagram, connections to alters eighteen years old or younger are valued significantly more than any other age group-two patterns not seen on any other platforms. Social media platforms individually generate between $53B and $215B in consumer surplus per year in the US alone. These results suggest social media generates significant value, local network effects drive a substantial fraction of that value and that these effects vary across platforms, consumers, and connections. |
| Date: | 2025–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2507.04545 |