|
on Discrete Choice Models |
| By: | Fanny Claise; Marielle Brunette |
| Abstract: | Natural events pose a real threat to forests around the world. Insurance contracts can help protect forest owners against these damaging events. However, there is considerable heterogeneity in terms of insurance adoption across countries. In France, for instance, the adoption rate is extremely low. In this article, we attempt to identify the characteristics of insurance contracts that influence forest owners’ demand for insurance against natural events. To this end, we employed a Discrete Choice Experiment methodology involving hypothetical forest insurance scenarios that varied according to the characteristics of the insurance contract such as the hazard(s) covered, the level of deductible, the duration, and the annual cost. The results, based on 317 responses from French private forest owners, demonstrate that some of the tested characteristics had a significant impact. Notably, forest owners were not willing to pay for storm insurance in addition to fire insurance. Conversely, they were willing to pay for insurance against the package including all hazards: fire, storm, drought and pathogens. |
| Keywords: | Forest Insurance; Discrete Choice Experiment; Contract; Logit; Willingness to pay (WTP); Contract |
| JEL: | B21 G22 Q23 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ulp:sbbeta:2025-42 |
| By: | Yeshwanth Cherapanamjeri; Constantinos Daskalakis; Gabriele Farina; Sobhan Mohammadpour |
| Abstract: | Random Utility Models (RUMs) are a classical framework for modeling user preferences and play a key role in reward modeling for Reinforcement Learning from Human Feedback (RLHF). However, a crucial shortcoming of many of these techniques is the Independence of Irrelevant Alternatives (IIA) assumption, which collapses \emph{all} human preferences to a universal underlying utility function, yielding a coarse approximation of the range of human preferences. On the other hand, statistical and computational guarantees for models avoiding this assumption are scarce. In this paper, we investigate the statistical and computational challenges of learning a \emph{correlated} probit model, a fundamental RUM that avoids the IIA assumption. First, we establish that the classical data collection paradigm of pairwise preference data is \emph{fundamentally insufficient} to learn correlational information, explaining the lack of statistical and computational guarantees in this setting. Next, we demonstrate that \emph{best-of-three} preference data provably overcomes these shortcomings, and devise a statistically and computationally efficient estimator with near-optimal performance. These results highlight the benefits of higher-order preference data in learning correlated utilities, allowing for more fine-grained modeling of human preferences. Finally, we validate these theoretical guarantees on several real-world datasets, demonstrating improved personalization of human preferences. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.15839 |
| By: | Stephane Hess; Sander van Cranenburgh |
| Abstract: | Models allowing for random heterogeneity, such as mixed logit and latent class, are generally observed to obtain superior model fit and yield detailed insights into unobserved preference heterogeneity. Using theoretical arguments and two case studies on revealed and stated choice data, this paper highlights that these advantages do not translate into any benefits in forecasting, whether looking at prediction performance or the recovery of market shares. The only exception arises when using conditional distributions in making predictions for the same individuals included in the estimation sample, which obviously precludes any out-of-sample forecasting. |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.09185 |
| By: | Sebastian Blesse; Klaus Gründler; Philipp Heil; Henning Hermes |
| Abstract: | Economic narratives are pervasive in the public discourse and can shape individual behavior. But so far we know very little about whether households actually demand and value narratives as information. We combine a comprehensive expert survey with a large-scale nationally representative household sample in the U.S. to examine the demand for economic narratives in a high-stakes environment of an unprecedentedly high recession probability. We document a substantial willingness to pay for economic narratives of more than 4 USD, which is higher than for numerical forecast information. The dominant motives for acquiring narratives are intrinsic, but a smaller share of participants also lists instrumental motives. Economic narratives improve respondents’ understanding of recession drivers and shape beliefs about the economy and spending, but exert only a minor impact on quantitative expectations. Our findings underscore the potential of narratives as a tool to improve economic understanding and to foster more informed decision-making. |
| Keywords: | narratives, experts, information acquisition, willingness to pay, expectation formation, belief formation, spending ontentions, recession |
| JEL: | D83 D84 D12 E32 E71 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12204 |
| By: | Tibor Szendrei (National Institute of Economic and Social Research); Arnab Bhattacharjee (Heriot-Watt University) |
| Abstract: | In models where the relationship between the outcome and the error term is linear, a residual can be computed by simply plugging in the estimated coeacients and computing the difference between observed and predicted values of the outcome variable. These residuals can then be used for many different purposes, for example: (a) evaluating assumptions of orthogonality of errors (like Rxed and random effects); (b) examining the entire shape of the error distribution; and (c) computation and inference on externalities such as network effects. However, this simple approach does not work when the model is nonlinear in outcomes and errors. Here different context-speciRc generalized residuals have been proposed, each having different properties for speciRc models. Note that for the canonical linear or nonlinear Gaussian regression model, the above construction is simply a scaled version of the partial derivative of the log-likelihood contribution of an individual observation with respect to the outcome variable. This suggests a general construction of generalized regression by perturbing the outcome variable and computing contrasts. This approach is closely related to Huber's inXuence function and can be routinely computed using Stata, for example, and also parallelized for large datasets. We propose this general construction of generalized residuals and evaluate its use in several contexts: (a) quantile regression and evaluation of conditional quantiles at the tails (for example, growth at risk); (b) computing errors distributions (for example, binary regression and random- effects models); and (c) computing network externalities in discrete choice and duration models. This delivers a uniRed approach with promising Rndings. |
| Date: | 2025–09–04 |
| URL: | https://d.repec.org/n?u=RePEc:boc:lsug25:17 |
| By: | Ernst Fehr; Julien Senn; Thomas Epper; Aljosha Henkel |
| Abstract: | In this registered report, we investigate (i) whether incentives affect subjects’ willingness to pay to increase, and to decrease the payoff of others, (ii) whether they affect the distribution of social preference types, and (iii) whether they affect the strength and the precision of individuals’ structurally estimated social preference parameters. Using an online experiment with a general population sample, we show that the use of monetary incentives, as well as the size of the stakes, have little impact on subjects’ modal choices (descriptive analysis), as well as for the distribution of qualitatively distinct preference types in the population (clustering analysis). However, monetary incentives affect quantitative measures of the strength and the precision of social preferences. Indeed, a structural analysis reveals that the preference elicitation with merely hypothetical stakes leads to an overestimation and a less precise measurement of social preferences. Together, these results highlight that incentivizing the elicitation of social preferences is most useful when interested in quantitative estimates. For researchers interested in identifying merely qualitative preferences types, however, hypothetical stakes might suffice. |
| Keywords: | Social preferences, altruism, inequality aversion, incentives |
| JEL: | C80 C90 D30 D63 |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:zur:econwp:482 |
| By: | Scholl, Lynn; Arellana, Julián; Cantillo, Víctor; Ojeda-Diaz, Alfredo J.; Oviedo, Daniel; Sabogal-Cardona, Orlando |
| Abstract: | Microtransit, or app-based collective transport, is a passenger transport service typically offered in medium-capacity vehicles using mobile phone apps. This service provides the advantages of public transport, allowing for more efficient use of vehicles, offering new opportunities to improve informal transit systems and reduce urban inequalities in Latin America and the Caribbean. This research examines how the level of service attributes, socioeconomic characteristics, and latent constructs (technological affinity, environmental attitudes, and security concerns in public transport) influence the willingness to use these services through two case studies in Mexico City, Mexico, and Barranquilla, Colombia. Data for this study comes from stated preference and perception surveys, which are commonly used in a psychometric and econometric approach to estimate integrated choice and latent variable models. The results indicate a high sensitivity to the price of the service. Attributes such as walking distance to access the service, travel time, service frequency, and schedule adherence reliability were also significant. There are substantial income differences in willingness to use microtransit services. Fare sensitivity is much higher among poorer segments of the population, affecting the potential of microtransit to address equity and inclusion issues in the cities studied. Of the latent constructs, only safety concerns about public transport were significant in the willingness to use microtransit services in both cities. When compared to men, women reported higher safety concerns and, as result, women have higher preference for microtransit services. Considering the results obtained from the modelling, sevearl policy considerations and actions are suggested to encourage the use of microtransit in the region and take advantage of its potential as a sustainable transport mode. |
| JEL: | O14 R42 R58 Z18 |
| Date: | 2025–08 |
| URL: | https://d.repec.org/n?u=RePEc:idb:brikps:14310 |
| By: | Siemroth, Christoph |
| Date: | 2025–10–15 |
| URL: | https://d.repec.org/n?u=RePEc:esx:essedp:41743 |
| By: | Peng Xu |
| Abstract: | This paper develops stochastic optimization problems for describing and analyzing behavioral investors with Markowitz Stochastic Dominance (MSD) preferences. Specifically, we establish dominance conditions in a discrete state-space to capture all reverse S-shaped MSD preferences as well as all subjective decision weights generated by inverse S-shaped probability weighting functions. We demonstrate that these dominance conditions can be admitted as linear constraints into the stochastic optimization problems to formulate computationally tractable mixed-integer linear programming (MILP) models. We then employ the developed MILP models in financial portfolio analysis and examine classic behavioral factors such as reference point and subjective probability distortion in behavioral investors' portfolio decisions. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.22896 |
| By: | Claudia Cerrone; Francesco Feri; Anita Gantner; Paolo Pin |
| Abstract: | This paper investigates whether the decoy effect - specifically the attraction effect - can foster cooperation in social networks. In a lab experiment, we show that introducing a dominated option increases the selection of the target choice, especially in early decisions. The effect is stronger in individual settings but persists in networks despite free-riding incentives, with variation depending on the decision-maker's strategic position. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.13887 |
| By: | Migrow, Dimitri (University of Edinburgh); Park, Hyungmin (University of Warwick); Squintani, Francesco (University of Warwick) |
| Abstract: | We study a leader’s choice of advisors, balancing political alignment, informational competence, and diversity of views. The leader can consult one or two advisors : one is politically aligned but less informed or shares potentially redundant information; the other is better informed but more biased. The leader’s optimal strategy can exhibit reversals. If both advisors are initially consulted, increasing the bias of the more biased advisor may cause the leader to exclude the aligned advisor to preserve truthfulness from the informed one. As bias rises further, the leader ultimately replaces the informed advisor if his bias becomes too large. When the leader is uncertain about the bias of the more informed advisor, increasing the chance of alignment can justify consulting both advisors. |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:wrk:wqapec:29 |
| By: | Agathe Pernoud; Frank Yang |
| Abstract: | A monopolist sells multiple goods to an uninformed buyer. The buyer chooses to learn any one-dimensional linear signal of their values for the goods, anticipating the seller's mechanism. The seller designs an optimal mechanism, anticipating the buyer's learning choice. In a generalized Gaussian environment, we show that every equilibrium has vertical learning where the buyer's posterior means are comonotonic, and every equilibrium is outcome-equivalent to nested bundling where the seller offers a menu of nested bundles. In equilibrium, the buyer learns more about a higher-tier good, resulting in a higher posterior variance on the log scale. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.16396 |
| By: | Alexander McFarlane |
| Abstract: | We derive a slippage-aware toxicity condition for on-chain liquidations executed via a constant-product automated market maker (CP-AMM). For a fixed (constant) liquidation incentive $i$, the familiar toxicity frontier $\nu |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2510.10171 |
| By: | Hoy, Christopher Alexander; Kim, Yeon Soo; Imtiaz, Saad; Rojas Mendez, Ana Maria; Meyer, Moritz; Canavire Bacarreza, Gustavo Javier; Kim, Lydia; Seitz, William Hutchins; Helmy, Imane; Uochi, Ikuko; Touray, Sering; Singh, Juni; Sjahrir, Bambang Suharnoko; Pape, Utz Johann; Fuchs Tarlovsky, Alan; Nguyen, Trang Van; Gencer, Defne; Lee, Min A; Sagesaka, Akiko; Contreras, Ivette |
| Abstract: | Public opposition is a major barrier to economic reforms, such as subsidy removal. Using multilayered, randomized survey experiments with 10, 000 respondents across ten surveys in five countries, this paper shows that opposition to energy price reforms is shaped more by design and communication than by cost. Around 70 percent of respondents strongly opposed a 100 percent immediate price increase, but resistance was nearly halved when reforms were phased in, targeted at high-energy consumers, or paired with compensation. Informational messages also reduced opposition by as much as halving the price increase. An expert prediction survey revealed systematic misunderstandings: specialists underestimated the influence of design features and greatly misperceived coping strategies and compensation preferences. These findings demonstrate that behavioral biases—such as present bias, loss aversion, and fairness heuristics—are as influential as economic costs in shaping people’s opposition to economic reforms, underscoring the importance of careful design and communication of politically sensitive reforms. |
| Date: | 2025–10–17 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wbrwps:11233 |
| By: | Charles Hodgson; Shilong Sun |
| Abstract: | How do vertically integrated firms' pricing and product provision decisions change with upstream and downstream competition? We answer this question in the context of the Chinese film industry. Theaters allocate significantly fewer showings to non-integrated films. This foreclosure effect is particularly pronounced in two scenarios: when an integrated theater faces limited spatial competition, and when an integrated film is similar to competing films. To measure welfare effects, we estimate a model of consumer preferences and theater showings choice using a novel method that combines standard demand data with film ratings data. Our results show that integrated theaters internalize a substantial portion of their upstream profits, driving foreclosure behavior that distorts showings. Counterfactual simulations show that vertical integration increases consumer welfare by 2.4% in the median market, but reduces consumer welfare in 7% of markets. The welfare effects of foreclosure vary with upstream competition between films and downstream competition between theaters, and we show that targeted antitrust policy that removes of integration based on measures of market competition can substantially increase welfare. |
| JEL: | L0 L13 L40 L42 L82 |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34390 |
| By: | Martin-Fuentes, Eva; Mellinas, Juan Pedro; Fernández, Cèsar; Font, Xavier (University of Surrey) |
| Abstract: | This study investigates whether consumer ratings of hotel eco-friendliness reflect actual sustainability performance. Using data from 6, 696 hotels in the world’s 100 leading destinations, we compared Expedia’s post-travel, customer-submitted eco-friendliness ratings with sustainability information reported on Booking.com, including both self-reported practices and third-party certifications. Support Vector Machine regression analysis shows that eco-friendliness ratings are explained almost entirely by overall guest satisfaction, with sustainability indicators contributing little explanatory power. This suggests that ratings conflate general service impressions with perceptions of environmental responsibility, limiting their value as measures of sustainability performance. While plausible explanations such as response biases and the limited salience of many certified practices warrant further research, our findings provide robust evidence that single survey items on eco-friendliness should be interpreted with caution. For platforms and policymakers, the results highlight the need to make sustainability cues more visible and directly tied to the consumer experience if ratings are to support informed choice. |
| Date: | 2025–10–17 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:dyz5e_v1 |
| By: | Gregorio Curello; Ludvig Sinander; Mark Whitmeyer |
| Abstract: | We uncover a close link between outside options and risk attitude: when a decision-maker gains access to an outside option, her behaviour becomes less risk-averse, and conversely, any observed decrease of risk-aversion can be explained by an outside option having been made available. We characterise the comparative statics of risk-aversion, delineating how effective risk attitude (i.e. actual choice among risky prospects) varies with the outside option and with the decision-maker's 'true' risk attitude. We prove that outside options are special: among transformations of a decision problem, those that amount to adding an outside option are the only ones that always reduce risk-aversion. |
| Date: | 2025–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2509.14732 |
| By: | Jirjahn, Uwe (University of Trier); Rienzo, Cinzia (University of Brighton) |
| Abstract: | Previous studies on working from home (WFH) and employee well-being have produced conflicting results. We hypothesize that giving workers a choice over whether to use WFH plays a crucial role in the consequences for well-being. This has a series of testable implications for empirical work. Using panel data from the UK, our fixed effects estimates show that not only the actual use, but also the pure availability of WFH is linked with improved job-related and overall mental health. Not controlling for the pure availability of WFH implies that the positive influence of the actual use of WFH is underestimated. However, we find a positive link between the use of WFH and overall mental health only for the years before and after the pandemic. The link was negative during the COVID-19 crisis where WFH was largely enforced. Moreover, gender moderates the influence of WFH on mental health. For women, both the actual use and the pure availability of WFH are positively associated job-related and overall mental health. For men, we find a more mixed pattern where either only the pure availability or only the actual use has an influence on mental health. Men are more likely to over- or underrate the consequences of WFH than women. |
| Keywords: | pandemic, COVID-19, freedom of choice, remote work, mental well-being, gender |
| JEL: | I10 I31 J16 J22 M50 |
| Date: | 2025–10 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18187 |
| By: | David Vincent (David Vincent Econometrics) |
| Abstract: | This presentation introduces a new command, xtfelevel, which implements a Hausman-type test to assess whether controlling for Rxed effects at a more aggregate (group) level is suacient for consistently estimating the coeacients on unit-speciRc, time-varying variables in linear panel-data models where units are nested within groups. The command builds on Papke and Wooldridge (2023), who develop a test of the null hypothesis that the probability limits of the Rxed-effects estimators for a coeacient of interest are the same whether heterogeneity is controlled at the unit or group level. Rejection of the null suggests that unit-level Rxed-effects estimation is required. xtfelevel extends this framework by comparing the unit-level Rxed effects estimator with an IV estimator that allows the time-varying controls to be correlated with unit-level heterogeneity, while accounting for correlation between the variable of interest and group- level effects. This estimator yields results analogous to pooled OLS estimation of the Mundlak regression, where the time average of the variable of interest is Rrst partialed out from the time averages of the controls. Under the null, the estimator can often be more eacient than the unit-level Rxed effects estimator, especially when the variable of interest exhibits limited within-unit variation. This extension addresses a limitation in applying the usual Mundlak device to obtain more eacient estimates, as discussed by Wooldridge (2019). When the variable of interest is uncorrelated with the unit-level heterogeneity but is correlated with the time-varying controls that are themselves correlated with those effects, excluding its time mean to improve eaciency can lead to omitted variable bias. |
| Date: | 2025–09–04 |
| URL: | https://d.repec.org/n?u=RePEc:boc:lsug25:12 |
| By: | Kumar, Praachi (Maastricht Graduate School of Governance, RS: GSBE MGSoG); Martorano, Bruno (Maastricht Graduate School of Governance, RS: GSBE MGSoG) |
| Abstract: | This research investigates the impact of exposure to the social media platform Twitter on son-biased fertility preferences for women in India, using information from over a million Tweets, combined with Demographic and Health Survey (DHS) data on more than a million respondents. We apply an instrumental variables strategy based on a popular national Twitter campaign attributed to the retirement of Indian cricketer, Sachin Tendulkar. We find that exposure to Twitter decreases discriminatory preferences regarding the sex of the child, particularly reducing son preference. These changes in preferences are mainly explained by the fact that social media content helps challenge harmful cultural norms. Specifically, we adopt a qualitative approach supported by a custom fine-tuned sequence classifier based on a pre-trained multilingual transformer encoder (XLM-RoBERTa) to show that Twitter served as a platform where Indian users discussed topics related to children, where most messages about children were neutral or progressive. We further demonstrate that content matters by focussing on an online campaign called #SelfieWithDaughter, and illustrate that social media exposure was particularly effective in shaping preferences in districts where the #SelfieWithDaughter campaign was active. We extend the main analysis to men and find that exposure to Twitter also reduces son preference for this group. Further evidence on the behavioural effects of social media exposure suggests a favourable but less straightforward effect on nutritional outcomes for girls under the age of five. All reported results from the heterogeneity analyses confirm that Twitter reduced discrimination, although it was less impactful for women in North India and those in districts with a higher scheduled caste population. |
| JEL: | J16 J13 O33 O12 L82 |
| Date: | 2025–10–20 |
| URL: | https://d.repec.org/n?u=RePEc:unm:unumer:2025023 |