|
on Discrete Choice Models |
| By: | Gabriel Nova; Sander van Cranenburgh; Stephane Hess |
| Abstract: | Discrete Choice Modelling serves as a robust framework for modelling human choice behaviour across various disciplines. Building a choice model is a semi structured research process that involves a combination of a priori assumptions, behavioural theories, and statistical methods. This complex set of decisions, coupled with diverse workflows, can lead to substantial variability in model outcomes. To better understand these dynamics, we developed the Serious Choice Modelling Game, which simulates the real world modelling process and tracks modellers' decisions in real time using a stated preference dataset. Participants were asked to develop choice models to estimate Willingness to Pay values to inform policymakers about strategies for reducing noise pollution. The game recorded actions across multiple phases, including descriptive analysis, model specification, and outcome interpretation, allowing us to analyse both individual decisions and differences in modelling approaches. While our findings reveal a strong preference for using data visualisation tools in descriptive analysis, it also identifies gaps in missing values handling before model specification. We also found significant variation in the modelling approach, even when modellers were working with the same choice dataset. Despite the availability of more complex models, simpler models such as Multinomial Logit were often preferred, suggesting that modellers tend to avoid complexity when time and resources are limited. Participants who engaged in more comprehensive data exploration and iterative model comparison tended to achieve better model fit and parsimony, which demonstrate that the methodological choices made throughout the workflow have significant implications, particularly when modelling outcomes are used for policy formulation. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.01704 |
| By: | Somers, Melline (RS: GSBE other - not theme-related research, ROA / Health, skills and inequality); Stolp, Tom (RS: GSBE other - not theme-related research, ROA / Education and transition to work); Burato, Francesca; Groot, Wim (Maastricht Graduate School of Governance, RS: GSBE MGSoG, RS: CAPHRI - R2 - Creating Value-Based Health Care, Health Services Research); van Merode, Frits (Faculteit FHML Centraal, RS: CAPHRI - R2 - Creating Value-Based Health Care); Vooren, Melvin |
| Abstract: | The healthcare and education sectors suffer from shortages of nurses and teachers. Extending their working hours has often been proposed as a solution to this. In this study, we conduct a discrete choice experiment (DCE) in the Netherlands to elicit nurses’ and teachers’ preferences for different jobs and working conditions. We present both nurses and teachers with nine hypothetical choice sets, each consisting of two jobs that differ in seven observable job attributes. From the DCE, we infer workers’ willingness to pay for these different job characteristics. Moreover, we calculate how many additional hours workers would be willing to work if a specific workplace condition were met. We find that both nurses and teachers most negatively value high work pressure. Spending a lot of time on patient-related tasks is highly valued by nurses, followed by having more control over working hours. Next to work pressure, teachers place significant importance on receiving social support from both colleagues and managers. Nurses and teachers who work part-time require higher incentives to work additional hours compared to full-time workers. |
| JEL: | J45 J81 J20 J30 I10 I20 |
| Date: | 2024–11–14 |
| URL: | https://d.repec.org/n?u=RePEc:unm:umagsb:2024014 |
| By: | Ubi-Abai, Itoro |
| Abstract: | This study assessed households’ willingness to pay and payments for water services supplied by the Akwa Ibom Water Company Limited using the Heckman two-step analysis. Using the survey research design, these households comprised households that have access to 5.06% of water supply services from the Akwa Ibom Water Company Limited a la Ubi-Abai (2024); and households that live close to the water company but do not have access to their water services. A sample of 200 households was selected using the two-stage cluster probability sampling and the purposive non-probability sampling techniques. Data were obtained using structured questionnaires. The weighted Kappa and Cronbach Alpha coefficients showed that questions in the questionnaires were valid and reliable. The cluster analysis revealed that 38 households used water efficiently. Furthermore, the Heckman two-step analysis revealed that factors such as water use efficiency, water quality, income and family size influenced households’ willingness to pay and monthly water payment levels. |
| Keywords: | Akwa Ibom, Heckman, Household, Water, WTP |
| JEL: | D1 D11 D12 I31 |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:pra:mprapa:122512 |
| By: | Kremena Valkanova |
| Abstract: | We examine the effect of item arrangement on choices using a novel decision-making model based on the Markovian exploration of choice sets. This model is inspired by experimental evidence suggesting that the decision-making process involves sequential search through rapid stochastic pairwise comparisons. Our findings show that decision-makers following a reversible process are unaffected by item rearrangements, and further demonstrate that this property can be inferred from their choice behavior. Additionally, we provide a characterization of the class of Markovian models in which the agent makes all possible pairwise comparisons with positive probability. The intersection of reversible models and those allowing all pairwise comparisons is observationally equivalent to the well-known Luce model. Finally, we characterize the class of Markovian models for which the initial fixation does not impact the final choice and show that choice data reveals the existence and composition of consideration sets. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.22001 |
| By: | Carlos Alos Ferrer; Ernst Fehr; Michele Garagnani |
| Abstract: | Transitivity is perhaps the most fundamental axiom in economic models of choice. The empirical literature has regularly documented violations of transitivity, but these violations pose little problem if they are simply a result of somewhat-noisy decision making and not a reflection of the deterministic part of individuals’ preferences. However, what if transitivity violations reflect genuinely nontransitive preferences? And how can we separate nontransitive preferences from noise-generated transitivity violations–a problem that so far appears unresolved? To tackle these fundamental questions, we develop a theoretical framework which allows for nontransitive choices and behavioral noise. We then derive a non-parametric method which uses response times and choice frequencies to distinguish genuine (and potentially nontransitive) preferences from noise. We apply this method to two different datasets, demonstrating that a substantial proportion of transitivity violations reflect genuinely nontransitive preferences. These violations cannot be accounted for by any model using transitive preferences and noisy choices. |
| Keywords: | Transitivity, Stochastic choice, Preference Revelation |
| JEL: | D01 D81 D87 D91 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:lan:wpaper:413755644 |
| By: | Federico Echenique; Quitz\'e Valenzuela-Stookey |
| Abstract: | Harsanyi (1955) showed that the only way to aggregate individual preferences into a social preference which satisfies certain desirable properties is ``utilitarianism'', whereby the social utility function is a weighted average of individual utilities. This representation forms the basis for welfare analysis in most applied work. We argue, however, that welfare analysis based on Harsanyi's version of utilitarianism may overlook important distributional considerations. We therefore introduce a notion of utilitarianism for discrete-choice settings which applies to \textit{social choice functions}, which describe the actions of society, rather than social welfare functions which describe society's preferences (as in Harsanyi). We characterize a representation of utilitarian social choice, and show that it provides a foundation for a family of \textit{distributional welfare measures} based on quantiles of the distribution of individual welfare effects, rather than averages. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.01315 |
| By: | Elaine Kelly (Institute for Fiscal Studies); Isabel Stockton (Institute for Fiscal Studies) |
| Date: | 2024–04–24 |
| URL: | https://d.repec.org/n?u=RePEc:ifs:ifsewp:24/11 |
| By: | Christopher D. Walker |
| Abstract: | This paper presents a Bayesian inference framework for a linear index threshold-crossing binary choice model that satisfies a median independence restriction. The key idea is that the model is observationally equivalent to a probit model with nonparametric heteroskedasticity. Consequently, Gibbs sampling techniques from Albert and Chib (1993) and Chib and Greenberg (2013) lead to a computationally attractive Bayesian inference procedure in which a Gaussian process forms a conditionally conjugate prior for the natural logarithm of the skedastic function. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.17153 |
| By: | Mohiuddin, Hossain; Fitch-Polse, Dillon T. PhD |
| Abstract: | To understand the extent to which micromobility services such as bike-share and scooter-share are enabling car-light lifestyles by replacing driving, we explore the trip-chaining patterns of micromobility users. We use travel diary data collected from micromobility users in 48 cities across the US. Our analysis incorporated 15, 985 trip chains from 1, 157 survey participants who provided at least seven days of travel diary data, and an imputed dataset of 35, 623 trip chains from 1, 838 participants from the same survey. Our analysis of both datasets shows that a considerable portion of car owners are leaving their cars at home when using micromobility. This suggests that, for a subset of users, micromobility can form part of a car-free or car-light day of travel, despite having a car available. Trip chains with less frequent car use are composed of a variety of different modes in combination with micromobility. Micromobility services are supportive of complex trip chains that include both work and non-work trips with reduced reliance on cars. The use of micromobility services tends to entirely replace shorter car trips on shorter-length trip chains. Our findings show the importance of considering the chain of trips rather than individual trips to understand the sustainability potential of micromobility services. The policy implications of these findings are improving methods of travel behavior analysis of shared mobility services. |
| Keywords: | Social and Behavioral Sciences, Micromobility, shared mobility, trip chaining, mode choice, travel surveys |
| Date: | 2024–09–01 |
| URL: | https://d.repec.org/n?u=RePEc:cdl:itsdav:qt2r66k788 |
| By: | Takeshi Fukasawa |
| Abstract: | This study proposes a simple procedure to compute Efficient Pseudo Likelihood (EPL) estimator proposed by Dearing and Blevins (2024) for estimating dynamic discrete games, without computing Jacobians of equilibrium constraints. EPL estimator is efficient, convergent, and computationally fast. However, the original algorithm requires deriving and coding the Jacobians, which are cumbersome and prone to coding mistakes especially when considering complicated models. The current study proposes to avoid the computation of Jacobians by combining the ideas of numerical derivatives (for computing Jacobian-vector products) and the Krylov method (for solving linear equations). It shows good computational performance of the proposed method by numerical experiments. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.20029 |
| By: | Yu, Shubin |
| Abstract: | This study explores the efficacy of large language models (LLMs) in short-text topic modeling, comparing their performance with human evaluation and Latent Dirichlet Allocation (LDA). In Study 1, we analyzed a dataset on chatbot anthropomorphism using human evaluation, LDA, and two LLMs (GPT-4 and Claude). Results showed that LLMs produced topic classifications similar to human analysis, outperforming LDA for short texts. In Study 2, we investigated the impact of sample size and LLM choice on topic modeling consistency using a COVID-19 vaccine hesitancy dataset. Findings revealed high consistency (80-90%) across various sample sizes, with even a 5% sample achieving 90% consistency. Comparison of three LLMs (Gemini Pro 1.5, GPT-4o, and Claude 3.5 Sonnet) showed comparable performance, with two models achieving 90% consistency. This research demonstrates that LLMs can effectively perform short-text topic modeling in medical informatics, offering a promising alternative to traditional methods. The high consistency with small sample sizes suggests potential for improved efficiency in research. However, variations in performance highlight the importance of model selection and the need for human supervision in topic modeling tasks. |
| Date: | 2024–11–01 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:mqk3r |
| By: | Sandro Ambuehl; Heidi C. Thysen |
| Abstract: | Good decision-making requires understanding the causal impact of our actions. Often, we only have access to correlational data that could stem from multiple causal mechanisms with divergent implications for choice. Our experiments comprehensively characterize choice when subjects face conflicting causal interpretations of such data. Behavior primarily reflects three types: following interpretations that make attractive promises, choosing cautiously, and assessing the fit of interpretations to the data. We characterize properties of interpretations that obscure bad fit to subjects. Preferences for more complex models are more common than those reflecting Occam’s razor. Implications extend to the Causal Narratives and Model Persuasion literatures. |
| Keywords: | Decision-making, causal mechanisms, causal narratives, model persuasion, causal interpretations |
| JEL: | C91 D01 D83 |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:zur:econwp:458 |
| By: | Ming Li; Zhentao Shi; Yapeng Zheng |
| Abstract: | This paper studies estimation and inference in a dyadic network formation model with observed covariates, unobserved heterogeneity, and nontransferable utilities. With the presence of the high dimensional fixed effects, the maximum likelihood estimator is numerically difficult to compute and suffers from the incidental parameter bias. We propose an easy-to-compute one-step estimator for the homophily parameter of interest, which is further refined to achieve $\sqrt{N}$-consistency via split-network jackknife and efficiency by the bootstrap aggregating (bagging) technique. We establish consistency for the estimator of the fixed effects and prove asymptotic normality for the unconditional average partial effects. Simulation studies show that our method works well with finite samples, and an empirical application using the risk-sharing data from Nyakatoke highlights the importance of employing proper statistical inferential procedures. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.23852 |
| By: | Chad Brown |
| Abstract: | I consider inference in a partially linear regression model under stationary $\beta$-mixing data after first stage deep neural network (DNN) estimation. Using the DNN results of Brown (2024), I show that the estimator for the finite dimensional parameter, constructed using DNN-estimated nuisance components, achieves $\sqrt{n}$-consistency and asymptotic normality. By avoiding sample splitting, I address one of the key challenges in applying machine learning techniques to econometric models with dependent data. In a future version of this work, I plan to extend these results to obtain general conditions for semiparametric inference after DNN estimation of nuisance components, which will allow for considerations such as more efficient estimation procedures, and instrumental variable settings. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.22574 |
| By: | Casey, Katherine (Stanford U); Glennerster, Rachel (U of Chicago) |
| Abstract: | Why are there few debates in low-information elections where they have the greatest potential to inform vote choices? Consistent with weak incentives to reveal their quality or make policy commitments, we find only a quarter of Parliamentary candidates in Sierra Leone privately volunteer to debate. Publicizing their choices through guaranteed dissemination platforms allows voters to punish those who abstain and sharply increases participation. Randomly improving platform quality induces frontrunners to join. We document high voter willingness to pay to access debates and private sector interest in disseminating them, confirming that candidate reluctance and not market viability is the main barrier. |
| JEL: | D72 O12 |
| Date: | 2024–08 |
| URL: | https://d.repec.org/n?u=RePEc:ecl:stabus:4178 |
| By: | Mitchell Linegar; Betsy Sinclair; Sander van der Linden; R. Michael Alvarez |
| Abstract: | Large Language Models (LLMs) can assist in the prebunking of election misinformation. Using results from a preregistered two-wave experimental study of 4, 293 U.S. registered voters conducted in August 2024, we show that LLM-assisted prebunking significantly reduced belief in specific election myths, with these effects persisting for at least one week. Confidence in election integrity was also increased post-treatment. Notably, the effect was consistent across partisan lines, even when controlling for demographic and attitudinal factors like conspiratorial thinking. LLM-assisted prebunking is a promising tool for rapidly responding to changing election misinformation narratives. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.19202 |
| By: | Andr\'as Telcs; Marcell T. Kurbucz; Antal Jakov\'ac |
| Abstract: | Temporally evolving systems are typically modeled by dynamic equations. A key challenge in accurate modeling is understanding the causal relationships between subsystems, as well as identifying the presence and influence of unobserved hidden drivers on the observed dynamics. This paper presents a unified method capable of identifying fundamental causal relationships between pairs of systems, whether deterministic or stochastic. Notably, the method also uncovers hidden common causes beyond the observed variables. By analyzing the degrees of freedom in the system, our approach provides a more comprehensive understanding of both causal influence and hidden confounders. This unified framework is validated through theoretical models and simulations, demonstrating its robustness and potential for broader application. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2410.19469 |
| By: | Alice Pizzo (Copenhagen Business School); Christina Gravert (Department of Economics, University of Copenhagen); Jan M. Bauer (Copenhagen Business School); Lucia Reisch (University of Cambridge, Judge Business School) |
| Abstract: | We examine the impact of a carbon tax on consumer choices via a large-scale online randomized controlled trial. Higher taxes generally reduce the demand for high-carbon goods. Compared to an import tax, a carbon tax reduces demand when the tax is zero (i.e., announced but not levied) but shows relatively higher demand for high-carbon goods when a positive tax is introduced. This contradiction of basic price theory is entirely driven by climate-concerned consumers. Our findings suggest that carbon taxes can crowd out climate concerns, leading to important implications for policy. |
| Keywords: | Behavioral response; Carbon pricing; Climate change; Experiment; Moral licensing. |
| JEL: | Q58 C90 D03 D90 Q50 Q51 |
| Date: | 2024–11–13 |
| URL: | https://d.repec.org/n?u=RePEc:kud:kucebi:2416 |
| By: | Carlos Alós-Ferrer; Ernst Fehr; Helga Fehr-Duda; Michele Garagnani |
| Abstract: | Recent contributions suggest that the empirical evidence for the common ratio effect could be explained as noise instead of underlying preferences under “common assumptions.” We revisit this argument using a more general method which allows to unambiguously dis- tinguish noise from preferences nonparametrically and with less stringent assumptions. The results are independent of the assumed behavioral model or how noise affects choices. Ap- plying this method to new experimental data we show that there is a systematic preference for the common ratio and the common consequence effects which cannot be explained by noise. |
| JEL: | C91 D81 D91 |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:zur:econwp:459 |
| By: | Limor Hatsor; Ronen Bar-El |
| Abstract: | An alternative to the dependence on traditional student loans may offer a viable relief from the tremendous burden that those loans usually incur. This article establishes that it is desirable for governmental intervention to grant students 'more choice' in their funding decisions by allowing them to have portfolios, mixtures of different types of loans. To emphasize this point, a model is presented of a situation where students invest in higher education while facing uncertainty about their individual earning potential. The model reveals that when students are allowed to have portfolios of loans, some of them indeed take the opportunity and diversify their loans, benefiting themselves, but also improving the loan terms of other students. Therefore, when governments organize student loans, they should consider providing students with more choice in their funding decisions. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.05506 |