|
on Discrete Choice Models |
| By: | Ecenur Oguz; Robert L. Bray |
| Abstract: | We develop the first general-purpose estimator for infinite-horizon dynamic discrete choice models whose estimation problem, after pre-computation, is unencumbered by large systems of linear equations -- either imposed as constraints, or embedded in the objective function. Our unnested fixed point (UFXP) and optimal unnested fixed point (OUFXP) estimators exploit a dual representation of Bellman's equation to separate the utility parameters from the dynamic programming fixed point. We establish the consistency and asymptotic normality of UFXP and OUFXP, as well as the efficiency of the latter. Our estimators enable researchers to model utility functions non-parametrically via flexible neural-network approximations. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.09736 |
| By: | Beth Forys; Paul Hindsley; O. Ashton Morgan; John C. Whitehead |
| Abstract: | Harmful algal blooms (HABs) reduce the amenity quality and perceived safety of coastal and freshwater resources. Events trigger avoidance behavior by residents and visitors and contribute to broader economic losses. Managers need economic welfare measures that correspond to the intensity categories used in public advisories. We quantify how advisory-defined HAB risk alters the expected value of outdoor recreation trips in South Florida for two hazards: cyanobacteria (blue-green algae; microcystins) and red tide (Karenia brevis). We administer a split-sample contingent valuation survey in four waves (Dynata; April 2024 to March 2025; n = 4, 135; 12, 405 trip decisions). Respondents first identified the destination of their next South Florida trip from 11 regions and then evaluated randomized trip-cost increases and HAB intensity scenarios aligned with Florida Fish and Wildlife Conservation Commission red tide tiers and U.S. EPA microcystin benchmarks. In our preferred model, we address hypothetical bias using a stacked logit model that calibrates for choice certainty and stated attribute non-attendance. In the most conservative specification, per-travel-party willingness to pay for an overnight trip is about US$1, 250 under no advisory, falls to roughly US$560Ð660 under low-intensity advisories, and drops to about US$200Ð330 under medium to high HAB intensity. These changes imply avoidance values of approximately US$600Ð700 per trip at low intensity and US$940Ð1, 040 at medium to high intensity, with stated trip taking probabilities near 50% at medium/high risk. Holding trip counts fixed (valuing observed trips only) and excluding substitution, a back-of-the-envelope calculation for Lee County, FL visitation over a 60-day event window suggests order-of-magnitude recreational welfare losses of US$17Ð58 million for cyanobacteria and US$60Ð103 million for red tide. These intensity-specific estimates provide transferable inputs for benefitÐcost analysis of South Florida water-management operations and nutrient-reduction policies. Key Words: harmful algal blooms, contingent valuation method, hypothetical bias, attribute non-attendance, choice certainty, willingness to pay |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:apl:wpaper:26-05 |
| By: | Hua Wang; Yuhan Deng; Donald S. Kenkel; Alan D. Mathios; Sen Zeng |
| Abstract: | A growing body of economic research explores the impacts of U.S. e-cigarette regulations on consumer tobacco choices, but less is known about e-cigarette regulation in China, the world’s largest tobacco market. We study China’s ban of flavored e-cigarettes. The ban of all flavors in e-cigarettes other than tobacco was part of a comprehensive package of regulatory policies adopted in 2022. We collected stated preference data through two discrete choice experiments conducted in 2021 and 2023, with about 600 subjects each. All subjects were adult current smokers. In the experiments, subjects made hypothetical choices between cigarettes, e-cigarettes, and quitting. Product prices and the attributes of e-cigarettes were experimentally varied, allowing us to identify the impact of flavor availability on stated preferences. We use the data to estimate conditional logit models and to predict the impact of the flavor ban and other policies. The empirical results suggest that a ban of flavored e-cigarettes decreases stated preferences for e-cigarettes but also has the unintended consequence to increase stated preferences for cigarettes. Despite the predicted decrease in e-cigarette choices, the predicted choice share of flavored e-cigarettes when they are illegal but loosely enforced is 53% of the predicted share when legal. This large illegal share is consistent with anecdotal evidence and with the evidence from our 2023 background survey that flavored e-cigarettes remain popular after the ban although fewer vapers reported getting their e-cigarettes from specialty or general retailers. |
| JEL: | I12 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:35048 |
| By: | Avidit Acharya; Jens Hainmueller; Yiqing Xu |
| Abstract: | Conjoint experiments randomize multidimensional profiles, offering a powerful design for recovering structural preference parameters -- including marginal rates of substitution, willingness to pay, and the distribution of preferences across a population. Yet the dominant approach in political science has focused on nonparametric causal estimands that do not leverage this potential. We propose a structural approach that embeds a deep neural network within a random utility logit model, allowing preference parameters to vary as a fully flexible function of respondent characteristics. The neural network addresses the concern that a parametric specification may not capture the true data generating process, while double/debiased machine learning provides valid inference on average preference parameters. We apply our method to three prominent conjoint studies and find rich preference heterogeneity masked by reduced-form averages: a near-zero gender effect coexists with 83% preferring female candidates, opposition to undemocratic behavior is near-universal but varies sharply in intensity, and progressive tax preferences cut across every partisan subgroup. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.10845 |
| By: | Nan Liu; Yanbo Liu; Yuya Sasaki; Yuanyuan Wan |
| Abstract: | The maximum score method (Manski, 1975, 1985) is a powerful approach for binary choice models, yet it is known to face both practical and theoretical challenges. In particular, the estimator converges at a slower-than-root-$n$ rate to a nonstandard limiting distribution. We investigate conditions under which strictly concave surrogate score functions can be employed to achieve identification through a smooth criterion function. This criterion enables root-$n$ convergence to a normal limiting distribution. While the conditions to guarantee these desired properties are nontrivial, we characterize them in terms of primitive conditions. Extensive simulation studies support, the root-$n$ convergence rate, the asymptotic normality, and the validity of the standard inference methods. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.13399 |
| By: | Harold D. Chiang; Ahnaf Rafi |
| Abstract: | The maximum score estimator of Manski (1975) provides an elegant approach to estimate slope coefficient in binary choice models without requiring parametric assumptions on the error distribution. However, under i.i.d. sampling, it admits a non-Gaussian limiting distribution and exhibits cube-root asymptotics, which complicates statistical inference. We show that, under multiway dependence, the maximum score estimator attains asymptotic normality at a parametric rate. We obtain this surprising result through the development of a general M-estimation theory that accommodates non-smooth objective functions under multiway dependence. We further propose and establish the validity of a bootstrap procedure for inference. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.10232 |
| By: | Markus Lill (LMU Munich); Nastasia Gallitz (LMU Munich); Lucas Stich (University of Wuerzburg); Martin Spann (LMU Munich) |
| Abstract: | Platform endorsement badges (e.g., Amazon's Choice) are widely believed to shape consumer decisions, yet their effectiveness in complex retail environments---where endorsements compete with rankings, ratings, and other signals---remains not well understood. This article examines Amazon's product-level endorsements using a multi-method approach combining (1) a 50-day large-scale audit of more than 200, 000 search results spanning over 90, 000 products and (2) a lab-in-the-field experiment that manipulates badge visibility and placement in consumers' natural shopping context. The audit reveals that endorsements are rare (~1.3% of products) and disproportionately assigned to products with lower prices, higher ratings, and those sold or fulfilled by Amazon; receiving a badge is associated with greater search visibility and improved sales performance. The experiment shows that displaying the badge tends to increase click-through and add-to-cart likelihoods, whereas reassigning or masking it tends to reduce these behaviors; however, these effects---while economically meaningful---are estimated with uncertainty, consistent with a multi-cue environment in which endorsement competes with other signals such as search rank and Prime eligibility. Together, the findings indicate that platform endorsement badges shape consumer search and choice behavior even in information-rich retail settings. Implications are discussed for platform design, seller strategy, and regulatory oversight. |
| Keywords: | platform endorsements; consumer decision-making; digtial platforms; e-commerce experimentation; |
| JEL: | D12 D83 L86 M31 |
| Date: | 2026–04–01 |
| URL: | https://d.repec.org/n?u=RePEc:rco:dpaper:569 |
| By: | Anke Becker (Harvard Business School); Christina Borner (LMU Munich); Thomas Dohmen (University of Bonn); Armin Falk (University of Bonn); David Huffman (Cornell University); Uwe Sunde (LMU Munich) |
| Abstract: | A growing body of empirical research has developed measures of economic preferences related to risk taking and intertemporal choice. This research has documented pronounced heterogeneity in preferences across and within societies, and also provided evidence that these differences are culturally transmitted. This chapter discusses existing data sets that allow for a comparable measurement across the globe, takes stock of commonalities and differences in approaches, and presents an extended synthetic cross-country data set that combines information from existing data sets. The analysis then establishes various empirical regularities, such as broadly similar patterns of heterogeneity across the globe, revealed by the different datasets, but also some systematic divergences by measurement approach, and substantial correlations of economic preferences with country-aggregate and individual-level outcomes and traits. We also briefly discuss international data sets measuring social preferences, and end with an outlook on avenues for future research. |
| Keywords: | willingness to take risks; patience; |
| JEL: | D1 |
| Date: | 2026–03–31 |
| URL: | https://d.repec.org/n?u=RePEc:rco:dpaper:568 |
| By: | Lixiong Li |
| Abstract: | Empirical researchers increasingly use upstream machine-learning (ML) methods to construct proxies for latent target variables from complex, unstructured data. A naive plug-in use of such proxies in downstream econometric models, however, can lead to biased estimation and invalid inference. This paper develops a framework for partial identification and inference in general moment models with ML-generated proxies. Our approach does not require restrictive assumptions on the upstream ML procedure, such as consistency or known convergence rates, nor does it require a complete validation sample containing all variables used in the downstream analysis. Instead, we assume access to two datasets: a downstream sample containing observed covariates and the proxy, and an auxiliary validation sample containing joint observations on the proxy and its target variable. We treat the proxy as a linking variable between these two samples, rather than as a literal noisy substitute for the latent target variable. Building on this idea, we develop a sharp identification strategy based on an unconditional optimal transport characterization and an inference procedure that controls asymptotic size using analytical critical values without resampling. Monte Carlo simulations show reliable size control and informative confidence sets across a range of predictive-accuracy scenarios. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.10770 |