[Truncated abstract] Air pollution is increasingly perceived to be a serious intangible threat to humanity, with air quality continuing to deteriorate in most urban areas. The main sources of inner city pollution are motor vehicles, which generate emissions from the tail pipe as well as by evaporation. These contain toxic gaseous components which have adverse health effects. The major components are carbon monoxide (CO), nitrogen dioxide (NO2), nitric oxide (NO), sulphur dioxide (SO2), particulates (PM10), and volatile organic compounds (VOC). CO and oxides of nitrogen (NOx) are major emissions from cars. This study focuses on pollutant concentration in Perth city and has sought to develop measures to improve air quality. To estimate concentrations, the study develops air pollution models for CO and NOx; on the basis of the model estimates, effective policy is devised to improve the air quality by managing travel to the city. Two peaks, due to traffic, are observed in hourly CO and NOx concentrations. Unlike traffic, however, the morning peak does not reach the level of the afternoon peak. The reasons for this divergence are assessed and quantified. Separate causal models of hourly concentrations of CO and NOx explain their fluctuations accurately. They take account of the complex effects of the urban street canyon and winds in the city. The angle of incidence of the wind has significant impact on pollution level; a wind flow from the south-west increases pollution and wind from the north-east decreases it. The models have been shown to be equivalent to engineering and scientific models in estimating emission rate in the context of street canyons. However the study models are much more precise in the Perth context. ... The models are used to calculate the marginal effects for all attributes and elasticity for fuel price. In almost all attributes the non-work group is more responsive than the work group. Finally, the SP model results are integrated into an econometric model for the purpose of prediction. The travel behaviour prediction is used to estimate the policy impact on air quality. The benefit from the air quality improvement is reported in terms of life saved. The estimated relationships between probability of death and air pollution determines the number of lives that could be saved under various policy scenarios. A ratio of benefits to the financial and perceived sacrifices by drivers is calculated to compare the effectiveness of the suggested policies. A car size charge policy was found to be the most cost effective measure to ameliorate the environmental impact of cars in Perth, with a morning peak entry time charge being almost as cost effective. The study demonstrates the need for appropriate modelling of air pollution and travel behaviour. It brings together analytical methods at three levels of causality, vehicle to air pollution, charge to travel response, and air pollution to health.

[pt] Esse estudo investiga em que medida o desenvolvimento e introdução de novas aeronaves moldam a estrutura da rede das companhias aéreas. Argumento que aeronaves modernas são mais eficientes e adequadas para operar voos entre cidades menores e menos centrais, favorecendo assim o serviço em mais mercados na periferia da rede. Com dados sobre o tráfego aéreo dos Estados Unidos, utilizo um arcabouço de escolha discreta para modelar as decisões de entrada das companhias e a subsequente escolha de aeronave em cada mercado. Experimentos contrafactuais mostram que, caso o desenvolvimento de tecnologia tivesse cessado em 1999, a rede de tráfego aéreo como um todo estaria mais centralizada, a maioria das companhias estariam operando redes mais centradas em torno de hubs, alcançando menos cidades e servindo menos mercados. / [en] This paper studies to what extent the development of new aircraft shapes airlines network structure. I argue that modern aircraft are more efficient and well suited to operate flights between smaller and less central cities, hence favoring the service of more markets in the periphery of the network. Using U.S. air traffic data, I employ a discrete choice framework to model airlines entry decisions and the subsequent aircraft choice to each market. Counterfactual experiments show that had aircraft technology ceased to improve in 1999, the air traffic network as a whole would be more centralized, airlines would be operating more hub-centered networks, reaching fewer cities, and serving fewer markets.

The study examines the job market of «urban youth» aged between 18 and 35 without higher education (university or technical) based on household surveys (Encuesta Nacional de Hogares – Enaho) and specialized surveys (Encuesta Nacional de Habilidades – Enhab; Encuesta de Transición de la Escuela al Trabajo – Entrans). The results show, firstly, supply constraints in the period 2007- 2012, which coincides with the massification of higher education over the last ten years. Secondly, it was found that students from public schools are more liable to complete high school with maxi- mum educational attainment. Thirdly, it was estimated that urban youths aged 15 to 29 cited in the Entrans 2012—and who received job training in the year prior to the survey—are 4.1 times more likely to obtain «adequate employment» in terms of pay, adequate contracts, and health insurance. Finally, a case is made for strengthening technical and specialized skills taught at school. / El estudio examina el mercado laboral de «jóvenes urbanos» entre 18 y 35 años de edad sin estudios superiores (universitarios o técnicos) a partir de Encuestas de Hogares (Enaho) y encuestas especializadas (Enhab, Entrans). Los resultados muestran, en primer lugar, una contracción de la oferta en el período 2007-2012 que coincide con la masificación de la educación superior de los últimos diez años. Segundo, se halló que los estudiantes de escuelas públicas son más propensos a alcanzar un máximo nivel educativo de secundaria completa. Tercero, se estimó que los jóvenes urbanos de 15 a 29 años reportados en la Entrans 2012 y que recibieron capacitaciones laborales durante el año anterior a la encuesta tienen 4,1 veces mayor probabilidad de obtener «empleados adecuados» —en términos de remuneraciones, contratos adecuados y seguro de salud. Finalmente, se propone potenciar las capacitaciones de tipo técnico y de especialización impartidas durante la escuela.

Les préférences des ménages français pour les travaux de rénovation énergétique dans le logement ont été peu étudiées jusqu’à présent. Or, elles sont la clé d’une politique énergétique efficace et réaliste pour le secteur résidentiel, tant au niveau des aides proposées, qu’au niveau des objectifs à réaliser. Les ménages sont-ils intéressés par des travaux à fort potentiel d’économie d’énergie ? Combien sont-ils prêt à payer et est-ce que c’est suffisant pour couvrir les coûts réels ? Combien de ménages n’ont pas le choix de leur mode de chauffage et quel est leur profil ?La thèse s'attache à répondre à ces questions de façon empirique à l’aide de modèles de choix discrets. Il s’agit, dans un premier temps, d’analyser les contraintes techniques, sociodémographiques et spatiales que rencontrent les ménages et qui entravent les choix énergétiques dans le logement. Une méthodologie de détection des ménages contraints a été proposée, puis appliquée au choix du chauffage en 2006 et 2013. Dans un deuxième temps, les préférences des ménages pour les équipements de chauffage et les travaux de rénovation ont été analysées. Cette deuxième partie utilise une enquête expérimentale qui a été réalisée spécialement pour ce travail de recherche auprès de 2000 ménages. Ce travail a permis de comprendre les préférences des ménages, leur hétérogénéité, et d’en déduire les consentements à payer, les taux d’actualisation implicites et les différents effets économiques. / Little research has been done to date on French households’ preferences for energy retrofits in the dwelling. However, the economic role of these choices is crucial for implementing an effective and realistic energy policy in the residential sector, both in terms of proposed aid and targets. Are households interested in retrofits with high energy-savings potential? How much are they willing to pay and is it sufficient to cover the up-front costs? How many households have no choice of heating system, and what is their profile?This work sets out to answer these questions empirically using discrete choice models. It starts by analyzing the technical, socio-demographic and spatial constraints that face households and create barriers to domestic energy choices. A method to detect household constraint is devised and applied to the choice of space heating system in 2006 and 2013. The second stage analyzes household preferences for heating equipment and retrofits. This second part is based on a discrete choice experiment specially carried out for this research among 2,000 households. This work sheds light on households’ preferences and heterogeneity, as well as their willingness to pay, the implicit discount rate and other economic impacts.

Submitted by FLAVIO MORAES (flafmoraes@yahoo.com.br) on 2010-08-19T14:21:41Z No. of bitstreams: 1 Dissertacao_Flavio_Moraes.pdf: 462004 bytes, checksum: 123425d1cbdd74b47091293eaf1f296d (MD5) / Approved for entry into archive by Andrea Virginio Machado(andrea.machado@fgv.br) on 2010-08-19T16:53:50Z (GMT) No. of bitstreams: 1 Dissertacao_Flavio_Moraes.pdf: 462004 bytes, checksum: 123425d1cbdd74b47091293eaf1f296d (MD5) / Made available in DSpace on 2010-08-20T14:46:06Z (GMT). No. of bitstreams: 1 Dissertacao_Flavio_Moraes.pdf: 462004 bytes, checksum: 123425d1cbdd74b47091293eaf1f296d (MD5) Previous issue date: 2008-07-18 / The purpose of this paper is to discuss in details the development of estimation methods of demand and supply parameters in di§erentiated products markets. The techniques presented consider explicitly the endogeneity of prices and can be applied to di§erent types of industries. The market demands system is derived from discrete choice models describing the behaviour of the consumer. This system is then combined with hypothesis about the cost functions and the behaviour of price determination by the Örms to generate equilibrium prices and quantities. The parameters to be estimated are those that determine the marginal costs of the Örms and the distribution of the consumersí tastes. This distribution determine elasticities and these, combined with the marginal cost and a hypothesis of Nash equilibrium, determine equilibrium prices. These elasticities and cost parameters play a central role in analysis of descriptive / Este trabalho tem por objetivo discutir detalhadamente o desenvolvimento de métodos de estimação de parâmetros de demanda e oferta em mercados de produtos diferenciados. As técnicas apresentadas consideram explicitamente a endogeneidade dos preços e podem ser aplicadas a diferentes tipos de indústrias. O sistema de demandas de mercado é derivado a partir de modelos de escolha discreta descrevendo o comportamento do consumidor. Esse sistema é então combinado com hipóteses sobre as funções custo e sobre o comportamento de determinação dos preços por parte das firmas para gerar preços e quantidades de equilíbrio. Os parâmetros a ser estimados são os que determinam os custos marginais das firmas e a distribuição dos gostos dos consumidores. Essa distribuição determina elasticidades e estas, combinadas com o custo marginal e com uma hipótese de equilíbrio de Nash na determinação de preços, determinam preços de equilíbrio. Essas elasticidades e parâmetros de custo desempenham um papel central em análises de questões descritivas e de mudanças no ambiente do mercado sob análise.

Cette thèse a pour enjeu de traiter des paradoxes étudiés en théorie du choix social.Le paradoxe d'Ostrogorski sur deux axes programmatiques a été traité, notamment sa probabilité de réalisation par l'ajout d'un critère discriminant sur les axes au moment de réaliser le choix de l'électeur : une formule de calcul exacte a été mise au point pour des valeurs de population finies afin de mesurer son occurrence pour différents effectifs, et une borne maximale émerge autours de 0,085.Parmi, les différentes anomalies étudiées en théorie du choix social affectant le fonctionnement des démocraties, le paradoxe du référendum occupe une place particulière du fait de son observation assez récurrente dans l'histoire électorale récente.L'un des enjeux de cette thèse a été de déterminer une méthode utilisable pour mesurer précisément sa probabilité d'occurrence dans des conditions précises de taille du corps électoral et de découpage.Il a été notamment recherché un moyen de comparer sa fréquence selon le nombre de circonscriptions retenu.Une formule a ainsi été déterminée pour des découpage du corps électoral en 3, 5, 7 et 9 circonscriptions de taille homogène.Un second résultat de la thèse sur le même paradoxe a été d'abolir l'hypothèse d'homogénéité parfaite des effectifs des circonscriptions pour mesurer l'effet de leur variation sur la probabilité de conflit pour un découpage en 3 circonscriptions.Des pistes ultérieures de recherche ont également explorées, en particulier la possibilité d'abolir partiellement l'hypothèse de culture neutre avec un découpage en 3 circonscriptions.Il a également été procédé à un état des lieux des types d'architecture institutionnelle, dont une classification globale en quatre catégories a été établie.Il a été tenté de déterminer leur poids dans les conflits de pouvoirs observés dans certains pays, en ayant notamment recours à des résultats obtenus grâce au paradoxe du référendum. / This thesis has aimed issues to deal with paradoxes studied in social choice theory.The Ostrogorski paradox with two programmatic axes was treated, including its achievement by adding a distinguishing criterion on the axes to realize the voter choice: an exact formula has been developed for a finite population to measure its occurrence for different numbers, and a effective maximum bound has emerged around 0.085.Among the various anomalies studied in social choice theory in the functioning of democracy, the referendum paradox holds a special place because of its fairly recurrent observation in recent electoral history.One of the stake of this thesis was to determine a suitable method to accurately measure its probability of occurrence in precise terms of size of the electorate and cutting.It was particularly sought a way to compare its frequency depending on the number of selected districts.A formula has been determined for cutting the electorate in 3, 5, 7 and 9 homogeneous size constituencies.A second result of the thesis on the same paradox was to relax the perfect homogeneity assumption on the constituencies size to measure the effect of their variation on the likelihood of conflict for a division into 3 districts.Subsequent research directions have also explored the possibility to partially abolish the assumption of impartial culture with a division into three districts.An inventory has been also conducted of the institutional architecture types.A comprehensive four-category classification was established, and we have tried to determine their weight in conflicts of powers observed in some countries,in particular using results deduced from the referendum paradox.

Made available in DSpace on 2016-06-02T20:00:12Z (GMT). No. of bitstreams: 1 DissMAA.pdf: 804171 bytes, checksum: a9e4a03249fb28379176634688b6e566 (MD5) Previous issue date: 2005-08-29 / Financiadora de Estudos e Projetos / The objective of this dissertation is to verify if there is a relationship between the characteristics of the urban form and option for walking trips. For this relationship to be studied, the first step was a bibliographic review about the urban form characteristics that might influence the option for walking. Among the methodologies that could be used to evaluate the modal option, a logit-type Discrete Choice Model, was chosen. The data used for the calibration of the model were collected in São Carlos, SP, a Brazilian medium-sized city with around 200 thousand inhabitants. The analyses were developed at census tract level, with a Geographic Information System (TransCAD). Two logit-type models were calibrated. The first one (basic) includes only a socio-economic variable (automobile availability) and the trip length. The second model (expanded) includes, besides the variables of the basic model, information about the characteristic of the urban form in the trip origin sectors. The comparison of the results obtained with the two models, allowed the evaluation of the marginal influence of urban form variables on the mode choice. In general, the expanded model was statistically better then the basic one. The results suggest that urban form characteristics do influence the individuals´ option for walking. / O objetivo desta pesquisa é verificar se existe relação entre as características da forma urbana e a opção dos indivíduos pelas viagens a pé. Para que essa relação pudesse ser estudada foi feito, inicialmente, um levantamento bibliográfico sobre as características da forma urbana que influenciam na opção pelo modo a pé. Dentre as metodologias mais usuais para avaliar o relacionamento entre as características da forma urbana e a opção pelo modo de transporte, foi selecionado o Modelo Comportamental ou de Escolha Discreta do tipo Logit. Os dados necessários à calibração do modelo foram coletados na cidade de São Carlos, SP, uma cidade de porte médio com cerca de 200 mil habitantes. As análises necessárias foram realizadas em nível de setor censitário e implementadas com o auxilio de um Sistema de Informações Geográficas. Foram calibrados dois modelos do tipo Logit. O primeiro modelo (básico) inclui apenas uma variável socioeconômica (disponibilidade de automóvel) e o comprimento da viagem a ser realizada. O segundo modelo (expandido) inclui, além das variáveis do modelo básico, informações sobre as características do meio físico urbano nos setores de origem das viagens. A comparação dos resultados obtidos com os modelos básico e expandido permitiu avaliar o efeito da inclusão das variáveis da forma urbana na expressão da utilidade do modo a pé. De maneira geral, o modelo expandido se mostrou estatisticamente melhor que o modelo básico. Os resultados permitem inferir que as características da forma urbana influenciam o comportamento dos indivíduos ao optar pelo modo a pé para realização de suas viagens.

AgÃncia Nacional do PetrÃleo / A utilizaÃÃo de modelos de escolha discreta à um mÃtodo eficaz que retrata o comportamento dos consumidores em diversos mercados. Sua aplicaÃÃo tem sido amplamente difundida na literatura para retratar a realidade de mercados de produtos e serviÃos no setor de transportes. Quando sÃo necessÃrios estudos de previsÃo de demanda, o modo mais adequado consiste na utilizaÃÃo conjunta de dados de preferÃncia declarada (PD) e preferÃncia revelada (PR). A combinaÃÃo destes dados fornece modelos estatisticamente mais consistentes do que aqueles estimados com dados puros de PD ou de PR. Dessa forma, o presente trabalho aplica uma metodologia baseada em modelos de escolha discreta com insumo de dados de PD e PR, chamada de GNVPREV, para analisar as preferÃncias dos usuÃrios de combustÃveis, enfocando o gÃs natural veicular - GNV, dentro de um contexto competitivo de um mercado de energÃticos veiculares. Esta anÃlise foi restrita aos usuÃrios de veÃculos leves - automÃveis, camionetas e caminhonetes - que utilizam a gasolina, o Ãlcool ou o prÃprio GNV. A metodologia GNVPREV foi aplicada para uma Ãrea de estudo constituÃda por uma parte do distrito sede do municÃpio de Caucaia, situado na RegiÃo Metropolitana de Fortaleza, Estado do CearÃ. O levantamento de dados a partir de questionÃrios de preferÃncia declarada e revelada, elaborados previamente, forneceu insumos para estimativas de funÃÃes de utilidade e obtenÃÃo de parÃmetros de elasticidade de demanda, trade-off entre alternativas e cenÃrios de previsÃo de demanda. Os resultados obtidos foram satisfatÃrios, dentro das limitaÃÃes dos dados primÃrios e secundÃrios, e confirmaram um melhor desempenho do modelo quando estimado com dados conjuntos de PD e PR.

Utilizou-se um modelo de escolha discreta (MED) para avaliar as preferências e disposição-a-pagar (DAP) dos pacientes por diferentes atributos de risco-benefício do tratamento insulínico, entre eles a via de administração da insulina. Através de uma revisão da literatura, consulta com especialistas, e do desenvolvimento de um estudo qualitativo utilizando as técnicas de entrevistas individuais e grupos focais, os atributos (e níveis) mais importantes do tratamento insulínico foram identificados, sob o ponto de vista dos pacientes. Os atributos incluídos no MED foram: controle da glicemia de jejum, número de episódios de hipoglicemia, ganho de peso, via de administração para as insulinas de ação longa e rápida, e custo do tratamento. Pares de opções de tratamentos insulínicos hipotéticos contendo diferentes níveis dos atributos foram apresentados aos pacientes com DM1 ou DM2, e lhes foi solicitado que, para cada cenário, eles escolhessem a alternativa de sua preferência. Dados demográficos, nível sócio-econômico (NSE) e informações relacionadas ao diabetes também foram coletados. Para a análise dos dados utilizou-se um modelo logit condicional de regressão e modelos segmentados foram posteriormente utilizados para a análise das sub-populações. Um total de 274 questionários foram incluídos na análise final dos dados. A idade média (± DP) dos participantes foi de 56.7 ± 12.98 anos, e 53% eram homens. Quarenta e nove por cento dos participantes eram usuários de insulina e 47 eram portadores de DM1. O tratamento insulínico ideal, sob o ponto de vista dos pacientes, resultaria em um melhor controle glicêmico, menos reações adversas, menor custo, e seria administrado por via oral. Houve uma forte preferência e uma DAP mais elevada por um melhor controle glicêmico, seguido pelos atributos de risco ganho de peso e episódios de hipoglicemia. Surpreendentemente, a via de administração da insulina foi o atributo menos valorizado. A estratificação social revelou que pacientes com alta renda anual familiar apresentaram uma DAP mais elevada por um melhor controle glicêmico e por menos reações adversas em relação aos grupos com rendas inferiores. Ainda, quanto mais alto o nível de renda, maior o desejo por uma insulina oral, enquanto a via inalada torna-se menos importante para os pacientes. A estratificação da amostra pelo uso de insulina e tipo de diabetes revelaram uma forte aversão pela via subcutânea pelos não-usuários de insulina e pacientes com DM2. Tais resultados sugerem a existência de uma importante barreira psicológica em se iniciar uso da insulina; no entanto, os resultados também revelam que os pacientes tendem a se acomodar com a via subcutânea uma vez iniciado o tratamento insulínico. Este estudo demonstra a importância que os pacientes com DM atribuem ao atributo controle glicêmico, e como suas preferências e DAP pelo tratamento insulínico variam entre as sub-populações. Especificamente, esforços devem ser realizados no sentido de vencer a barreira psicológica em se iniciar o uso da insulina, o que contribuirá para que se alcance um melhor controle glicêmico, através da melhor aderência do paciente ao tratamento, resultando em uma redução dos custos do DM e melhora na qualidade de vida dos pacientes. / We used a discrete choice experiment (DCE) to evaluate patients preferences for various attributes of insulin treatment, including route of insulin delivery. Through a review of the literature, expert consultation, and a qualitative descriptive study using individual interviews and focus group techniques, the attributes (and levels) of diabetes treatment most important to patients were identified. The attributes included in the DCE were: glucose control, frequency of hypoglycaemic events, weight gain, route of administration for the long-acting and the short-acting insulin, and out-of-pocket cost. Patients with type 1 or type 2 diabetes were presented with pairs of hypothetical insulin therapy profiles (i.e. choice sets) with different levels of the attributes and were asked to choose the treatment option they preferred. Sociodemographic data and diabetes medication were also collected. Data were analysed using conditional logit regression and segmented models were also developed for the analysis of subgroups. A Two hundred and seventy four questionnaires were completed. The mean age (±SD) of participants was 56.7 ± 12.98 years, and 53% were men. Forty-nine percent of participants were insulin users, and 47 had type 1 diabetes. Overall, patients ideal insulin treatment would provide better glucose control, result in fewer adverse reactions, have the lowest cost, and be administered orally. There was a strong preference and highest mean WTP for glucose control followed by the risk attributes weight gain and hypoglycaemic events. Surprisingly, route of insulin administration was the least valued attribute. Stratification of the sample revealed that patients with higher incomes had a significant higher WTP for better glucose control and less adverse reactions compared to lower income groups. Moreover, the highest the income, the stronger the preference for an oral insulin, while inhaled insulin becomes less important for patients. Segmented models by insulin use and type of diabetes indicated that insulin naïve and type 2 diabetics had a greater aversion to the subcutaneous route. These findings suggest that there may be an important mental barrier to initiating insulin therapy; however, patients tend to accommodate to subcutaneous administration once they initiate therapy. This study illustrates the importance that patients with diabetes place on glucose control and how preferences for insulin therapy vary between subgroups. Specifically, efforts need to be made to overcome the mental barrier to initiating insulin therapy, which may lead to improved control, through improved compliance and ultimately reduce the financial burden of the disease and improve patients quality of life.

This thesis addresses UK students’ university choice using discrete choice methods and micro-data obtained from Higher Education Statistical Agency for graduates between 2006 and 2010. The thesis consists of three chapters with each addressing a different aspect of students’ choice. The studies are intended to provide policy-makers and other decision-makers with valuable information that will help them to implement strategies and policies for better higher education. Some work in the literature has been dedicated to students’ university choice. This thesis explores this body of work and builds on it, extends it and improves what is previously known in the literature. The aim of the first chapter is to investigate what affects students’ university choice. It contributes to the literature by establishing the best method to do so. Two models are used: the standard conditional logit and conditional logit with, what is called in this paper, alternative specific constants. Conditional logit with alternative specific constants improves on conditional logit twofold: it deals with unobserved university characteristics and improves the model fit. The results show that the probability of attending a university decreases with an increase in tuition fees and distance between students’ home and the university, and decreases in students’ socio-economic status. The second chapter further investigates the importance of distance on students’ university choice and it contributes to the literature by calculating the willingness to pay of students for distance to university. The chosen models are estimated for different socio-economic group of students separately. This methodology allows for meaningful comparison between socio-economic groups and produces more reliable estimates due to the fact that it accommodates for different unobserved characteristics of universities for different groups of students. The results show that students with the highest socio-economic status are not affected or have a positive utility of distance. The willingness to pay of other socio-economic groups are mixed and depend on the university characteristics used in the model. The third chapter focuses on students’ attitudes towards costs and benefits of university degree by calculating the discount rate of future income using marginal utility of graduate income and tuition fees. In addition, the chapter shows how use of consideration sets of universities for each student improves the model fit. The results show that students have a normal discount rate around 1% without consideration sets. The discount rate becomes negative in all models apart from one, when consideration sets are used.