|
on Discrete Choice Models |
| By: | Anna Bartczak (University of Warsaw, Faculty of Economic Sciences); Wiktor Budziński (University of Warsaw, Faculty of Economic Sciences); Ulf Liebe (University of Warwick, Departement of Sociology); Jurgen Meyerhoff (Berlin School of Economics and Law) |
| Abstract: | In this paper, we investigate the effect of respondents’ attitudes concerning distributive justice in payments on their stated preferences for programmes reducing ambient air pollution in four cities in Poland. By combining two multi-factorial survey experiments, we propose a novel approach of incorporating justice attitudes into non-market valuation. In the first experiment – a factorial survey experiment (FSE) – we record justice attitudes towards payments. In the second experiment – a choice experiment (CE) – we elicit stated preferences for air pollution reduction programmes. As a modelling framework, we employ a hybrid choice model. The same respondents undertook both experiments in separate surveys one to two weeks apart, minimising the likelihood of biased estimates of the effect of justice attitudes on stated preferences. The results indicate a substantial effect of the justice attitude on the stated willingness to pay. The proposed approach could be used for joint modelling of justice attitudes and preferences in a wide range of fields, contributing further insights into their interactions. |
| Keywords: | air pollution, choice experiment, distributive justice attitude, factorial survey experiment, hybrid choice model, willingness to pay |
| JEL: | D63 I18 Q51 Q53 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:war:wpaper:2023-26&r=dcm |
| By: | Fu Ouyang; Thomas Tao Yang |
| Abstract: | We propose two approaches to estimate semiparametric discrete choice models for bundles. Our first approach is a kernel-weighted rank estimator based on a matching-based identification strategy. We establish its complete asymptotic properties and prove the validity of the nonparametric bootstrap for inference. We then introduce a new multi-index least absolute deviations (LAD) estimator as an alternative, of which the main advantage is its capacity to estimate preference parameters on both alternative- and agent-specific regressors. Both methods can account for arbitrary correlation in disturbances across choices, with the former also allowing for interpersonal heteroskedasticity. We also demonstrate that the identification strategy underlying these procedures can be extended naturally to panel data settings, producing an analogous localized maximum score estimator and a LAD estimator for estimating bundle choice models with fixed effects. We derive the limiting distribution of the former and verify the validity of the numerical bootstrap as an inference tool. All our proposed methods can be applied to general multi-index models. Monte Carlo experiments show that they perform well in finite samples. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.00013&r=dcm |
| By: | Marie-Laure Breuillé (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Institut Agro Dijon - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement); Julie Le Gallo (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Institut Agro Dijon - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement); Alexandra Verlhiac (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Institut Agro Dijon - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement) |
| Abstract: | Much has been written about the potential effect of the COVID-19 crisis on residential mobility. To explore its effects in France, we reconstruct flows of mobility intentions based on owner and buyer estimates on the platform MeilleursAgents from January 2019 to September 2021, and we analyze, using logit and nested logit models, how the pandemic has changed the probability that individuals from both urban and rural intend to relocate. Our results show that, after a time of shock during the first lockdown in spring 2020, the desire to migrate, either to rural municipalities or to other catchment areas, increased as the pandemic and the restrictive measures continued, and was particularly pronounced after the end of the third and last lockdown. |
| Keywords: | COVID‑19, platform data, residential location choice, discrete choice models, real estate |
| Date: | 2022–12–30 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-03910242&r=dcm |
| By: | Mai-Thi Ta (CIRED - Centre International de Recherche sur l'Environnement et le Développement - Cirad - Centre de Coopération Internationale en Recherche Agronomique pour le Développement - EHESS - École des hautes études en sciences sociales - AgroParisTech - ENPC - École des Ponts ParisTech - Université Paris-Saclay - CNRS - Centre National de la Recherche Scientifique); Léa Tardieu (UMR TETIS - Territoires, Environnement, Télédétection et Information Spatiale - Cirad - Centre de Coopération Internationale en Recherche Agronomique pour le Développement - AgroParisTech - CNRS - Centre National de la Recherche Scientifique - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement); Harold Levrel (CIRED - Centre International de Recherche sur l'Environnement et le Développement - Cirad - Centre de Coopération Internationale en Recherche Agronomique pour le Développement - EHESS - École des hautes études en sciences sociales - AgroParisTech - ENPC - École des Ponts ParisTech - Université Paris-Saclay - CNRS - Centre National de la Recherche Scientifique) |
| Abstract: | As the multiple benefits from exposure to urban green spaces (UGSs) are increasingly acknowledged, urban greening policies have become an important component of the urban political agenda. Most targeting strategies of future UGS development are based on the pursuit of an equal distribution of UGSs among residents. These strategies implicitly assume that the development of any type of UGS will have the same effect on citizens' well-being, provided that their access is guaranteed. This paper questions this assumption and addresses the demand side of urban greening policies by evaluating which UGS characteristics are sought by urban dwellers. We apply a travel time-based discrete choice experiment (DCE) to capture the trade-offs between the UGS characteristics (e.g., tree cover, size, presence of water, accessibility) and the travel time that citizens are willing to spend to reach a hypothetical UGS compared to a "stay at home" option. We discover that all the respondents have a disutility in choosing the "stay at home" option instead of a scenario of UGS development, especially when the UGS contains trees. This disutility is however much higher among outer suburb inhabitants living in municipalities with relatively lower urbanization levels and rent prices. Further, the global time budget dedicated to reach a UGS is much lower for inner-city residents compared to outer-suburb inhabitants. Inhabitants living in less urbanized areas place a higher value on a large UGS (> 1.5 hectares), while residents living in city centres do not seem to be influenced by this UGS characteristic. Our results suggest that strategies based on access criteria would benefit from being differentiated according to urbanization levels of cities, as the inhabitants of city centres value nearby and multiple UGSs but not necessarily large UGSs while the inhabitants of suburbs value larger UGSs, even when located farther away. urban green spaces-recreational services-urban greening policies-preference heterogeneity-choice experiment-green infrastructures |
| Abstract: | Les multiples avantages de l'exposition aux espaces verts urbains (EVU) étant de mieux en mieux reconnus, les politiques de renaturation sont devenues une composante importante de l'agenda politique urbain. La plupart des stratégies de ciblage des futurs EVU sont fondées sur la recherche d'une répartition égale des espaces verts urbains entre les résidents. Ces stratégies supposent implicitement que le développement de tout type d'EVU aura le même effet sur le bien-être des citoyens, à condition que leur accès soit garanti. Cet article interroge cette hypothèse en caractérisant la demande en EVU et en spécifiant les caractéristiques recherchées selon les profils sociodémographiques des habitants. Pour cela, nous avons réalisé une expérience de choix discrets basée sur des temps de trajets mesurant les arbitrages entre différents attributs constitutifs des EVU (couvert forestier, taille, forme, accessibilité) et le temps de trajet que les habitants accepteraient d'effectuer pour se rendre dans un espace vert fictif. Nous montrons que l'ensemble des résidents a une désutilité à choisir l'option "rester à la maison" plutôt qu'un scénario de développement d'un EVU, en particulier lorsque l'EVU contient des arbres. Cette désutilité est cependant beaucoup plus élevée chez les habitants des banlieues, vivant dans des municipalités à faibles taux d'urbanisation et à loyers modérées. Par ailleurs, le budget temps global consacré pour atteindre un EVU est beaucoup plus faible pour les habitants des centres-villes que pour ceux des banlieues. Enfin, les habitants des zones moins urbanisées accordent une plus grande valeur aux grands EVU (> 1, 5 hectares), tandis que les habitants des centres-villes ne semblent pas être influencés par cette caractéristique. Les résultats suggèrent que les stratégies basées sur des critères d'accès gagneraient à être différenciées en fonction du taux d'urbanisation des villes, car les habitants des centres-villes semblent mieux valoriser des EVU proches et nombreux mais pas nécessairement grands, tandis que les habitants des banlieues valorisent des EVU plus grands, même lorsqu'ils sont plus éloignés |
| Keywords: | urban green spaces, recreational services, urban greening policies, preference heterogeneity, choice experiment, green infrastructures, espaces verts urbains, services récréatifs, végétalisation, hétérogénéité des préférences, expérience de choix, infrastructures vertes |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-04210911&r=dcm |
| By: | Giovanni Compiani; Ilya Morozov; Stephan Seiler |
| Abstract: | We propose a demand estimation method that allows researchers to estimate substitution patterns from unstructured image and text data. We first employ a series of machine learning models to measure product similarity from products’ images and textual descriptions. We then estimate a nested logit model with product-pair specific nesting parameters that depend on the image and text similarities between products. Our framework does not require collecting product attributes for each category and can capture product similarity along dimensions that are hard to account for with observed attributes. We apply our method to a dataset describing the behavior of Amazon shoppers across several categories and show that incorporating texts and images in demand estimation helps us recover a flexible cross-price elasticity matrix. |
| Keywords: | demand estimation, unstructured data, computer vision, text models |
| JEL: | C10 C50 C81 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ceswps:_10695&r=dcm |
| By: | Ghislain Geniaux (ECODEVELOPPEMENT - Unité de recherche d'Écodéveloppement - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement) |
| Abstract: | In this working paper, I aim to establish a connection between the traditional mod- els of spatial econometrics and machine learning algorithms. The objective is to determine, within the context of big data, which variables should be incorporated into autoregressive nonlinear models and in what forms: linear, nonlinear, spatially varying, or with interactions with other variables. To address these questions, I propose an extension of boosting algorithms (Friedman, 2001; Buhlmann et al., 2007) to semi-parametric autoregressive models (SAR, SDM, SEM, and SARAR), formulated as additive models with smoothing splines functions. This adaptation primarily relies on estimating the spatial parameter using the Quasi-Maximum Like- lihood (QML) method, following the examples set by Basile and Gress (2004) and Su and Jin (2010). To simplify the calculation of the spatial multiplier, I propose two extensions. The first is based on the direct application of the Closed Form Estimator (CFE), recently proposed by Smirnov (2020). Additionally, I suggest a Flexible Instrumental Variable Approach/control function approach (Marra and Radice, 2010; Basile et al., 2014) for SAR models, which dynamically constructs the instruments based on the functioning of the functional gradient descent boosting algorithm. The proposed estimators can be easily extended to incorporate decision trees instead of smoothing splines, allowing for the identification of more complex variable interactions. For discrete choice models with spatial dependence, I extend the SAR probit model approximation method proposed by Martinetti and Geniaux (2018) to the nonlinear case using the boosting algorithm and smoothing splines. Using synthetic data, I study the finite sample properties of the proposed estimators for both Gaussian and probit cases. Finally, inspired by the work of Debarsy and LeSage (2018, 2022), I extend the Gaussian case of the nonlinear SAR model to a more complex spatial autoregressive multiplier, involving multiple spatial weight matrices. This extension helps determine the most geographically relevant spatial weight matrix. To illustrate the efficacy of functional gradient descent boosting for additive nonlinear spatial autoregressive models, I employ real data from a large dataset on house prices in France, assessing the out-sample accuracy. |
| Keywords: | Spatial Autoregressive model, gradient boosting, |
| Date: | 2023–05–25 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-04229868&r=dcm |
| By: | Nail Kashaev; Natalia Lazzati; Ruli Xiao |
| Abstract: | We develop a general model of discrete choice that incorporates peer effects in preferences and consideration sets. We characterize the equilibrium behavior and establish conditions under which all parts of the model can be recovered from a sequence of choices. We allow peers to affect only preferences, only consideration, or both. We exploit different types of variations to separate the peer effects in preferences and consideration sets. This allows us to recover the set (and type) of connections between the agents in the network. We then use this information to recover the random preferences and the attention mechanisms of each agent. These nonparametric identification results allow unrestricted heterogeneity across agents and do not rely on the variation of either covariates or the set of available options (or menus). We apply our results to model expansion decisions by coffee chains and find evidence of limited consideration. We simulate counterfactual predictions and show how limited consideration slows down competition. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2310.12272&r=dcm |
| By: | Martin Magris; Alexandros Iosifidis |
| Abstract: | The Bayesian estimation of GARCH-family models has been typically addressed through Monte Carlo sampling. Variational Inference is gaining popularity and attention as a robust approach for Bayesian inference in complex machine learning models; however, its adoption in econometrics and finance is limited. This paper discusses the extent to which Variational Inference constitutes a reliable and feasible alternative to Monte Carlo sampling for Bayesian inference in GARCH-like models. Through a large-scale experiment involving the constituents of the S&P 500 index, several Variational Inference optimizers, a variety of volatility models, and a case study, we show that Variational Inference is an attractive, remarkably well-calibrated, and competitive method for Bayesian learning. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2310.03435&r=dcm |
| By: | Christopher P. Chambers; Yusufcan Masatlioglu; Collin Raymond |
| Abstract: | Many models of economics assume that individuals distort objective probabilities. We propose a simple consistency condition on distortion functions, which we term distortion coherence, that ensures that the function commutes with conditioning on an event. We show that distortion coherence restricts belief distortions to have a particular function form: power-weighted distortions, where distorted beliefs are proportional to the original beliefs raised to a power and weighted by a state-specific value. We generalize our findings to allow for distortions of the probabilities assigned to both states and signals, which nests the functional forms widely used in studying probabilistic biases (e.g., Grether, 1980 and Benjamin, 2019). We show how coherent distorted beliefs are tightly related to several extant models of motivated beliefs: they are the outcome of maximizing anticipated expected utility subject to a generalized Kullback-Liebler cost of distortion. Moreover, in the domain of lottery choice, we link coherent distortions to explanations of non-expected utility like the Allais paradox: individuals who maximize subjective expected utility maximizers conditional on coherent distorted beliefs are equivalent to the weighted utility maximizers studied by Chew [1983]. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2310.09879&r=dcm |
| By: | Rylie Weaver |
| Abstract: | Ranked choice voting is vulnerable to monotonicity failure - a voting failure where a candidate is cost an election due to losing voter preference or granted an election due to gaining voter preference. Despite increasing use of ranked choice voting at the time of writing of this paper, the frequency of monotonicity failure is still a very open question. This paper builds on previous work to develop conditions which can be used to test if it's possible that monotonicity failure has happened in a 3-candidate ranked choice voting election. |
| Date: | 2023–09 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2310.12988&r=dcm |
| By: | Anne Margarian (Thünen Institute of Market Analysis); Cécile Détang-Dessendre (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Institut Agro Dijon - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement); Aleksandra Barczak (Département EcoSocio - Département Économie et Sciences Sociales pour l'Agriculture, l'Alimentation et l'Environnement - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement, CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Institut Agro Dijon - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement); Corinne Tanguy (CESAER - Centre d'Economie et de Sociologie Rurales Appliquées à l'Agriculture et aux Espaces Ruraux - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement - Institut Agro Dijon - Institut Agro - Institut national d'enseignement supérieur pour l'agriculture, l'alimentation et l'environnement) |
| Abstract: | Abstract Some rural locations in industrialized countries have experienced considerable employment growth in the last decades, while others suffer from depopulation and decline. The paper aims to contribute to the development of an evolutionary approach that allows for the identification of those often difficult-to-observe evolving factors that explain success and failure of rural locations. It also wants to show how the combined recognition of evolutionary labour market perspectives, the dynamic capability view of the firm, and human resource management (HRM) theories can serve the operationalisation of evolutionary explanations in this context. According to the derived model, apparent locational disadvantages might be compensated for by subtle, potentially self-enforcing labour market dynamics that generate opportunities for certain firms and industries. Empirically, the ideas are substantiated by means of a mediation model. The empirical analysis is based on latent class analysis and discrete choice models using data from an own survey of 200 food-processing firms in urban and rural locations of one German federal state. For these observations, our results support the idea that the exploitation of HRM opportunities may be more important for good performance in rural labour markets than the direct implementation of specific innovation modes. Investment in HRM allows rural firms in our sample to realise those gains in terms of innovation and growth offered by the creation of a stable and experienced workforce. Their focus on internal labour markets potentially generates external effects, which further encourages neighbouring firms to also invest in involved HRM measures. |
| Keywords: | Agglomeration advantages, Innovation, Human resource management, Mediation |
| Date: | 2022–07–05 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-03736538&r=dcm |