|
on Discrete Choice Models |
| By: | Rixt Bijker (KAW architects); Eveline van Leeuwen (Wageningen University & Research); Paul (P.R.) Koster (Vrije Universiteit Amsterdam) |
| Abstract: | This paper investigates how social interactions impact the decision to participate in one’s local environment. Existing work often reports correlations between social interactions and local participation, but it is unclear what the causal direction of this relationship is. A key contribution of this paper is that we are able to estimate the causal effect of social interactions on the decision to participate by systematically varying social attributes in a choice experiment. Based on a large-scale survey in one Dutch municipality we analyze 3894 choice observations of 435 respondents. Our sample includes respondents who currently participate and respondents who do not. We find that at the recruitment stage being asked by a friend or acquaintance significantly increases the chances to volunteer. We also find significant homophily effects in terms of age as well as for the characteristics of the group already participating. Financial incentives have significant negative impacts on the decision to participate. |
| Keywords: | Social interactions; Discrete choice experiments; Homophily; Volunteering |
| JEL: | C90 Z13 L30 R58 |
| Date: | 2019–01–11 |
| URL: | http://d.repec.org/n?u=RePEc:tin:wpaper:20190003&r=all |
| By: | Lukasz Grzybowski; Ambre Nicolle |
| Abstract: | In this paper, we use a unique dataset on switching between mobile handsets in a sample of about 8,623 subscribers using tariffs without handset subsidies from a single mobile operator on a monthly basis between July 2011 and December 2014. We estimate a discrete choice model in which we account for disutility from switching to different operating systems and handset brands and for unobserved time-persistent preferences for operating systems and brands. Our estimation results indicate the presence of significant inertia in the choices of operating systems and brands. We find that it is harder for consumers to switch from iOS to Android and other operating systems than from Android and other operating systems to iOS. Moreover, we find that there is significant time-persistent heterogeneity in preferences for different operating systems and brands, which also leads to state-dependent choices. We use our model to simulate market shares in the absence of switching costs and conclude that the market shares of Android and smaller operating systems would increase at the expense of the market share of iOS. |
| Keywords: | smartphones, consumer inertia, switching costs, mixed logit, iOS, Android |
| JEL: | L13 L50 L96 |
| Date: | 2018 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ceswps:_7434&r=all |
| By: | Berges, Miriam; Casellas, Karina; Pace Guerrero, Ignacio; Liseras, Natacha; Urquiza Jozami, Gonzalo; Echeverría, Lucía |
| Abstract: | Las enfermedades transmitidas por los alimentos (ETA) constituyen un importante problema de salud pública. El contexto actual de medios masivos de comunicación y exposición creciente a la información no garantiza consumidores mejor informados, capaces de discernir los riesgos y prevenirlos. Si bien los consumidores asegurarían que su grado de preocupación acerca de la seguridad de los alimentos que adquieren es muy alto, no siempre se verifican comportamientos de compra congruentes con esta declaración. Este trabajo indaga sobre las percepciones de los riesgos derivados del consumo de carne vacuna, en particular aquellos asociados a la contaminación bacteriológica como el Síndrome Urémico Hemolítico y la Escherichia Coli. El análisis se centró en los riesgos derivados del establecimiento de compra y sus características que permiten inferir inocuidad: las personas que despachan mercadería no estén en contacto con el dinero, las herramientas y utensilios empleados en la manipulación de la carne sean los apropiados y que los diferentes productos cárnicos se exhiban en heladeras debidamente separados y ordenados. A partir del empleo de Experimento de Elección o Choice Experiment (CE) que consiste en simular la elección de las características de la carnicería en la que el encuestado realizaría su hipotética compra, se estima la disposición a pagar por los diversos atributos relacionados con las prácticas de manipulación realizadas en el lugar de compra. Las elecciones de los consumidores se modelan a través de un Logit Multivariado (ML) que permite estimar la disposición a pagar (DAP) promedio por los distintos atributos de la carnicería. Sin embargo, las diferencias en el grado de información que poseen los consumidores modificará la utilidad de cada uno de estos atributos y por lo tanto su DAP, por lo que se diferencian las elecciones de acuerdo a su nivel de información. Tal como es de esperar, los individuos mejor informados sobre riesgos de contaminación están dispuestos a pagar entre un 16% y un 80% más por los atributos, mientras que los menos informados pagarían entre un 7% y un 35% menos por los mismos atributos. |
| Keywords: | Carne Vacuna; Comportamiento del Consumidor; Disposición a Pagar; Riesgo; Comercio Minorista; |
| Date: | 2018–09 |
| URL: | http://d.repec.org/n?u=RePEc:nmp:nuland:2978&r=all |
| By: | Beutel, Johannes; List, Sophia; von Schweinitz, Gregor |
| Abstract: | This paper compares the out-of-sample predictive performance of different early warning models for systemic banking crises using a sample of advanced economies covering the past 45 years. We compare a benchmark logit approach to several machine learning approaches recently proposed in the literature. We find that while machine learning methods often attain a very high in-sample fit, they are outperformed by the logit approach in recursive out-of-sample evaluations. This result is robust to the choice of performance measure, crisis definition, preference parameter, and sample length, as well as to using different sets of variables and data transformations. Thus, our paper suggests that further enhancements to machine learning early warning models are needed before they are able to offer a substantial value-added for predicting systemic banking crises. Conventional logit models appear to use the available information already fairly effciently, and would for instance have been able to predict the 2007/2008 financial crisis out-of-sample for many countries. In line with economic intuition, these models identify credit expansions, asset price booms and external imbalances as key predictors of systemic banking crises. |
| Keywords: | early warning system,logit,machine learning,systemic banking crises |
| JEL: | C35 C53 G01 |
| Date: | 2019 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:iwhdps:22019&r=all |
| By: | Alzola, Agustina; Lupín, Beatriz |
| Abstract: | Existe suficiente evidencia científica acerca de que una dieta sana y balanceada, junto con la adopción de hábitos saludables, son relevantes para mejorar la calidad de vida. Tanto el sobrepeso como la obesidad pueden ser perjudiciales para la salud dada su estrecha relación con enfermedades no transmisibles (ENT). A nivel individual, las mismas pueden ser disminuidas limitando el consumo de grasas totales, sodio y azúcares a favor de cereales integrales, legumbres, frutos secos y frutas y verduras frescas e intensificando la actividad física. Precisamente, las dietas y guías alimentarias y los organismos encargados de formular políticas referidas a la salud, recomiendan una ingesta de 400 g (5 porciones) de frutas y verduras por día. Si bien los especialistas coinciden en que un comportamiento saludable por parte de los jóvenes es clave para prevenir determinadas enfermedades, investigaciones de diversos países indican que, en general, éstos presentan hábitos alimentarios incorrectos, destacándose un consumo insuficiente de frutas y verduras. De esta manera, el interés del Trabajo se centra en el estudio cuantitativo de la alimentación de los jóvenes de Mar del Plata, con especial énfasis en el consumo de verduras. A tal fin, se estima un Modelo Logit Ordinal con datos provenientes de una submuestra de 120 jóvenes, de entre 18 y 29 años de edad, extraída de una encuesta sobre alimentación y hábitos saludables, relevada en dicha localidad, durante el año 2014, bajo un diseño probabilístico. La variable dependiente es la frecuencia de consumo semanal de verduras. Entre los resultados obtenidos, se destaca que el nivel socioeconómico, el sexo, el cuidado de la salud y la práctica regular de actividad física inciden en el consumo de verduras por parte de la población objetivo. |
| Keywords: | Consumo de Alimentos; Salud; Jóvenes; Modelo Logit; Mar del Plata; |
| Date: | 2018–09 |
| URL: | http://d.repec.org/n?u=RePEc:nmp:nuland:2975&r=all |
| By: | Ensslen, Axel; Gnann, Till; Jochem, Patrick; Plötz, Patrick; Dütschke, Elisabeth; Fichtner, Wolf |
| Abstract: | Plug-in electric vehicles are seen as a promising option to reduce oil dependency, greenhouse gas emissions, particulate matter pollution, nitrogen oxide emissions and noise caused by individual road transportation. But how is it possible to foster diffusion of plug-in electric vehicles? Our research focuses on the question whether e-mobility product service systems (i.e. plug-in electric vehicles, interconnected charging infrastructure as well as charging platform and additional services) are supportive to plug-in electric vehicle adoption in professional environments. Our user oriented techno-economic analysis of costs and benefits is based on empirical data originating from 109 organizational fleets participating in a field trial in south-west Germany with in total 327 plug-in electric vehicles and 181 charging points. The results show that organizations indicate a high willingness to pay for e-mobility product service systems. Organizations encounter non-monetary benefits, which on average overcompensate the current higher total cost of ownership of plug-in electric vehicles compared to internal combustion engine vehicles. However, the willingness to pay for e-mobility charging infrastructure and services alone is currently not sufficient to cover corresponding actual costs. The paper relates the interconnected charging infrastructure solutions under study to the development of the internet of things and smarter cities and draws implications on this development. |
| Keywords: | Electric mobility; electric vehicle; Smart city; Platform service; Business model; Product service system |
| JEL: | O33 R42 |
| Date: | 2018–05–21 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:91402&r=all |
| By: | LeSage, James P.; Chih, Yao-Yu; Vance, Colin |
| Abstract: | Focus is on efficient estimation of a dynamic space-time panel data model that incorporates spatial dependence, temporal dependence, as well as space-time covariance and can be implemented in large N and T situations, where N is the number of spatial units and T the number of time periods. Quasi-maximum likelihood (QML) estimation in cases involving large N and T poses computational challenges because optimizing the (log) likelihood requires: 1) evaluating the log-determinant of an NT x NT matrix that appears in the likelihood, 2) imposing stability restrictions on parameters reflecting space-time dynamics, as well as 3) simulations to produce an empirical distribution of the partial derivatives used to interpret model estimates that require numerous inversions of large matrices. We set forth a Markov Chain Monte Carlo (MCMC) estimation procedure capable of handling large problems, which we illustrate using a sample of T = 487 daily fuel prices for N = 12, 435 German gas stations, resulting in N x T over 6 million. The procedure produces estimates equivalent to those from QML and has the additional advantage of producing a Monte Carlo integrated estimate of the log-marginal likelihood, useful for purposes of model comparison. Our MCMC estimation procedure uses: 1) a Taylor series approximation to the logdeterminant based on traces of matrix products calculated prior to MCMC sampling, 2) block sampling of the spatiotemporal parameters, which allows imposition of the stability restrictions, and 3) a Metropolis-Hastings guided Monte Carlo integration of the logmarginal likelihood. We also provide an efficient approach to simulations needed to produce the empirical distribution of the partial derivatives for model interpretation. |
| Keywords: | dynamic panel models,spatial dependence,Markov Chain Monte Carlo estimation |
| JEL: | C23 D40 |
| Date: | 2018 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:rwirep:769&r=all |
| By: | Salomé Bakaloglou; Dorothée Charlier (ART-Dev - Acteurs, Ressources et Territoires dans le Développement - CIRAD - Centre de Coopération Internationale en Recherche Agronomique pour le Développement - UM3 - Université Paul-Valéry - Montpellier 3 - UPVD - Université de Perpignan Via Domitia - UM - Université de Montpellier - CNRS - Centre National de la Recherche Scientifique) |
| Abstract: | The aim of this research is to understand the impact of preference heterogeneity in explaining energy consumption in French homes. Using a discrete-continuous model and the conditional mixed-process estimator (CMP) enable us to address two potential endogeneities in residential energy consumption: energy prices and the choice of home energy characteristics. As a key contribution, we provide evidence that a preference for comfort over saving energy does have significant direct and indirect impacts on energy consumption (through the choice of dwelling), particularly for high-income households. Preferring comfort over economy or one additional degree of heating implies an average energy overconsumption of 10% and 7.8% respectively, up to 18% for high-income households. Our results strengthen the belief that household heterogeneity is an important factor in explaining energy consumption and could have meaningful implications for the design of public policy tools aimed at reducing energy consumption in the residential sector. |
| Keywords: | Residential energy consumption,Household preferences,Discrete-continuous choice method,Conditional mixed-process |
| Date: | 2018 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:halshs-01961638&r=all |
| By: | Lupín, Beatriz; Pérez, Stella Maris; Cincunegui, Carmen; Tedesco, Lorena |
| Abstract: | Un experimento de elección, como el Choice Modelling (CHM), permite analizar la valoración multiatributo que se realiza de un determinado producto. Los participantes enfrentan bloques de elección, con diversas alternativas del producto, descriptas por combinaciones de los niveles de los atributos, que deben elegir. Constituye un método de "preferencias declaradas", en situaciones simuladas de compra, permitiendo formular modelos econométricos más eficientes que con "preferencias reveladas". Su fundamentación conceptual es la Teoría del Consumo de Lancaster (1966) y el Modelo de Utilidad Aleatoria de Marschak (1960). Dado que en el Sudoeste Bonaerense (SOB) se produce un aceite de oliva (AO) de calidad diferenciada y a que los productores se encuentran evaluando la posibilidad de construir una Marca Colectiva Territorial, se desarrolló un CHM, en la Ciudad de Bahía Blanca, durante los meses de noviembre y diciembre del año 2017. El interés de este Trabajo se centra en presentar el diseño de un CHM , constituyendo un potencial aporte para aquellos que se encuentren estudiando la valoración de atributos de calidad de alimentos novedosos o con escasa presencia en el mercado. |
| Keywords: | Experimentos de Elección; Preferencias del Consumidor; Aceite de Oliva; |
| Date: | 2018–10 |
| URL: | http://d.repec.org/n?u=RePEc:nmp:nuland:2995&r=all |
| By: | Bernhard Ganglmair; Timothy Simcoe; Emanuele Tarantino |
| Abstract: | Motivated by a descriptive analysis of standards development within the Internet Engineering Task Force, we develop a dynamic discrete choice model of R&D that highlights the decision to continue or abandon a line of research. Our estimates imply that sixty percent of IETF proposals are publishable, but only one-third of those good ideas survive the review process. Increased attention and author experience are associated with faster learning. We simulate two counterfactual innovation policies: an R&D subsidy and a publication-prize. Subsidies have a larger impact on research output, though prizes perform better when accounting for researchers' opportunity costs. |
| Keywords: | Learning, Experimentation, Standardization, Dynamic Discrete Choice |
| JEL: | D83 O31 O32 |
| Date: | 2018–09 |
| URL: | http://d.repec.org/n?u=RePEc:bon:boncrc:crctr224_2018_041&r=all |
| By: | Lacaze, María Victoria; González, Julia |
| Abstract: | This paper evaluates the effect on market shares and consumer surplus of the introduction of a Good Agricultural Practices (GAP)-labeled product in the frozen fried potatoes (FFP) industry. We first estimate a model of household demand in Mar del Plata, Argentina, using scanner data and demographic information. We find that higher income individuals are more concerned about health and nutrition, and that younger and lower-income consumers are more price-sensitive. Then we postulate that a properly GAP-labeled FFP is available in the market, and we assess its effect by using the estimated utility function and prior information about consumers' declared willingness to pay (WTP) for sustainably produced potatoes. We find that the older the individual, the greater the influence of the hypothetical introduction of the GAP-labeled product; the relationship is less conclusive in the case of income. Finally, we predict the results of a greater consumer surplus extraction by fixing a higher price for the new product, and we calculate the maximum increase in the marginal cost that the firm would be able to afford if farmers charge higher prices for GAP fresh potatoes. |
| Keywords: | Buenas Prácticas Agrícolas; Alimentos Congelados; Papa; Disposición a Pagar; Modelo de Elección Discreta; |
| Date: | 2018–08 |
| URL: | http://d.repec.org/n?u=RePEc:nmp:nuland:2976&r=all |
| By: | Ryo Kato (Research Institute for Economics & Business Administration (RIEB), Kobe University, Japan); Takahiro Hoshino (Department of Economics, Keio University, Japan and RIKEN Center for Advanced Intelligence Project, Japan) |
| Abstract: | Issues regarding missing data are critical in observational and experimental research, as they induce loss of information and biased result. Recently, for datasets with mixed continuous and discrete variables in various study areas, multiple imputation by chained equation (MICE) has been more widely used, although MICE may yield severely biased estimates. We propose a new semiparametric Bayes multiple imputation approach that can deal with continuous and discrete variables. This enables us to overcome the shortcomings of multiple imputation by MICE; they must satisfy strong conditions (known as compatibility) to guarantee that obtained estimators are consistent. Our exhaustive simulation studies show thatthe coverage probability of 95 % interval calculated using MICE can be less than 1 %, while the MSE of the proposed one can be less than one-fiftieth. We also applied our method to the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, and the results are consistent with those of the previous research works that used panel data other than ADNI database, whereas the existing methods such as MICE, resulted in entirely inconsistent results. |
| Keywords: | Full conditional specification, Missing data, Multiple imputation, Probit stickbreaking process mixture, Semiparametric Bayes model |
| Date: | 2018–05 |
| URL: | http://d.repec.org/n?u=RePEc:kob:dpaper:dp2018-15&r=all |