Conselho Nacional de Desenvolvimento Científico e Tecnológico / Since the origins of modern finance theory, interdependence is an issue between assets as the search for diversification ends with a better portfolio. Many studies and many approaches have been given to answer to the questions of correlation, covariance and interdependence but the subprime crises and the almost meltdown of financial markets throughout the globe, proved that portfolios were not efficiently diversified. The concept of diversification and the risk measures did not prevent from hundreds of companies worldwide to go belly up. This work aims to give another approach to interdependence, the quantile regression, including a volatility measure of the American Market regressed with other international major players grouped by their geographic region. This work stars in 2000 and ends in 2012, collecting data through four major events: the Nasdaq burst, the 9/11, the Subprime and the European crises. Our findings are: countries geographically close to the US have display higher levels of interdependence; European countries are in majority also highly integrated with the US; and lastly but more importantly, integration is increasing. / Desde as origens das teorias em finanças se estuda a interdependência entre os ativos em busca dos benefícios de uma diversificação eficiente. Inúmeros estudos foram realizados e várias abordagens foram dadas ao problema da correlação, covariância e interdependência, mas a crise do Subprime e o consequente quase colapso dos mercados mundiais, provaram que os métodos utilizados até então não se mostraram eficientes, pois não foram capazes de evitar a evaporação de centenas de instituições financeiras, de diferentes mercados e extrema perda de valor de ativos dos mais diversificados portfólios e carteiras com diversos níveis de risco. Esse trabalho visa dar um novo enfoque ao estudo da interdependência, o da regressão quantílica com um componente de volatilidade do mercado americano regredido contra os retornos dos principais mercados mundiais, divididos por região geográfica, relacionados com o mercado americano através de seu principal índice, o S&P500, desde os anos de 2000 até 2012. Ao final, constata-se que os países geograficamente mais próximos dos Estados Unidos sofrem com maior integração, bem como a maioria e dos países do continente europeu, e principalmente que as integrações vem aumentando entre grande parte dos países estudados e o principal mercado mundial.

Cette thèse de doctorat est une première tentative d'examiner si les réglementations bancaires ont le même impact sur la stabilité et l'efficience des banques islamiques que sur celles des banques conventionnelles. Suite aux nouvelles recommandations de Bâle III, nous étudions l'impact des exigences minimales en matière de fonds propres, de liquidité et de levier financier sur la stabilité et l'efficience des banques islamiques comparativement aux banques conventionnelles. Une première étude exploratoire utilise l'analyse en composantes principales (ACP), les méthodes Logit et Probit et les régressions MCO pour montrer que les banques islamiques disposent d'un capital plus élevé, qu'elles sont plus liquides, plus profitables, mais moins stables que leurs homologues conventionnelles. Une deuxième étude empirique examine la stabilité des banques islamiques et utilise la régression quantile pour montrer que les banques islamiques sont moins stables que les banques classiques. L'étude prouve également que des exigences de fonds propres renforcées améliorent la stabilité des banques islamiques les plus petites et les plus liquides, tandis que le levier financier est négativement associé à la stabilité de ce type de banques. Des contraintes de liquidité plus fortes renforcent la stabilité des grandes banques islamiques alors que l'effet est inverse pour les petites banques. Enfin, nous examinons l'efficience des banques islamiques en utilisant la méthode d'enveloppement des données (DEA). Nous constatons que les banques islamiques sont plus efficientes que les banques conventionnelles. Nous trouvons aussi que des exigences de capital et de liquidité accrues pénalisent l'efficience des petites banques islamiques très liquides, alors que l'inverse est vrai pour le levier financier. Ces résultats montrent notamment qu'en matière de réglementation du capital pour les petites banques islamiques très liquides, un choix est à opérer entre une efficience accrue ou une stabilité renforcée. / This PhD dissertation is the first attempt to examine whether banking regulations have the same impact on the stability and the efficiency of Islamic than for conventional banks. We benefit of Basel III recommendations to investigate the impact of bank capital, liquidity and leverage requirements on the stability and the efficiency of Islamic banks compared to conventional banks. A first exploratory study uses Principal Component Analysis, Logit and Probit methods, and OLS regressions and shows that Islamic banks have higher capital, liquidity, and profitability, but that they are less stable than their conventional counterparts. A second empirical study examines the stability of Islamic banks using conditional quantile regressions and proves that Islamic banks are less stable than conventional banks. It also shows that higher capital and lower leverage improve the adjusted profits of small and highly liquid Islamic banks. Liquidity is positively associated with the stability of large Islamic banks while an opposite effect is detected when small Islamic banks are examined. Finally, we study the efficiency of Islamic banks using Data Envelopment Analysis (DEA) and find that Islamic banks are more efficient than conventional banks. We also find that higher capital and liquidity requirements penalize the efficiency of small and highly liquid Islamic banks, while the opposite is true for financial leverage. These results show that concerning capital requirements for small and highly liquid Islamic banks, a possible trade-off could be found between stability and efficiency.

Submitted by Viviane Lima da Cunha (viviane@biblioteca.ufpb.br) on 2017-07-06T14:30:18Z No. of bitstreams: 1 arquivototal.pdf: 6614647 bytes, checksum: 21e96f422787eaffc9d7176f0a15e007 (MD5) / Made available in DSpace on 2017-07-06T14:30:18Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 6614647 bytes, checksum: 21e96f422787eaffc9d7176f0a15e007 (MD5) Previous issue date: 2017-02-20 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / In this dissertation the beta regression models with variable dispersion and quantile regression are discussed. Therefore, an introduction was made with the objective of motivating its discussion in epidemiological studies, emphasizing the problematization around obesity. The application of these methods considered a real data set, obtained from public information sources, referring to adult obesity in the nations in the year 2014. After the descriptive analysis of the data it was verified that 50% of the nations present values of the proportion of obese adults greater than 0.20. In addition, viewing the obesity map by nation showed that the highest concentration of countries with the lowest obesity values is found in the continents of Asia and Africa. On the other hand, the highest concentrations of obese are found in the continents of America and Europe. Also, from the graphical analysis of the box-plot a possible difference in the proportions of obese adults between the continents of America and Europe with those of Africa and Asia was observed. After adjusting the beta and quantile regression models it was verified that the covariates average alcohol consumption in liters per person, percentage of insufficient physical activity and percentage of the population living in urban areas have a positive effect on the response variable. That is, individually such covariables tend to increase obesity values in the countries when the other covariables remain constant. In addition, the life expectancy variable in years presented a positive effect and was significant only for the variable regression beta regression model. Finally, analyzing the measures of prediction errors, it was verified that the estimates from the beta regression are more accurate when the mean square error and the total percentage error were evaluated. Therefore, for questions of predicting values for adult obesity in the nations in 2014, the beta regression model with variable dispersion was more suitable for this purpose. / Nesta dissertação são abordados os modelos de regressão beta com dispersão variável e de regressão quantílica. Para tanto, foi feita uma introdução com objetivo de motivar sua discussão em estudos epidemiológicos, enfatizando a problematização em torno da obesidade. A aplicação destes métodos considerou um conjunto de dados reais, obtidos a partir de fontes de informação pública, referente a obesidade adulta nas nações no ano de 2014. Após a análise descritiva dos dados verificou-se que 50% das nações apresentam valores da proporção de adultos obesos maiores do que 0.20. Além disso, visualizando o mapa da obesidade por nação constatou-se que a maior concentração de países com menores valores de obesidade encontra-se nos continentes da Ásia e África. Por outro lado, as maiores concentrações de obesos encontram-se nos continentes da América e Europa. Ainda, a partir da análise gráfica do box-plot foi observado uma possível diferença nas proporções de adultos obesos entre os continentes da América e Europa com os da África e Ásia. Após ajustar os modelos de regressão beta e quantílica verificou-se que as covariáveis consumo médio de álcool em litros por pessoa, porcentagem de atividade física insuficiente e porcentagem da população que vivem em áreas urbanas apresentam efeito positivo sobre a variável resposta. Ou seja, individualmente tais covariáveis tendem a aumentar os valores de obesidade nos países quando as demais covariáveis permanecem constantes. Além disso, a variável expectativa de vida em anos apresentou efeito positivo e foi significativa apenas para o modelo de regressão beta com dispersão variável. Por fim, analisando as medidas de erros de previsão verificou-se que as estimativas oriundas da regressão beta são mais precisas quando avaliado o erro quadrático médio e o erro percentual total. Portanto, para questões de predizer valores referentes a obesidade adulta nas nações em 2014 o modelo de regressão beta com dispersão variável se mostrou mais adequado para tal propósito.

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / The role of Brazilian cities had great changes over the last twenty-six. The process of decentralization allowed with the 1988 Constitution aimed the financial and political strengthening of the states and especially the municipalities in relation to the federal government. Thus, given the cities greater autonomy, as became 'federative'. Areas such as health and education have gained proper attribution of the local governments and increase the administrative burden, thereby increasing the financial and cities now have to raise more to fulfill them. The administration of a municipality is what organizes and prepares it for development. The need for a management organized so that the population is well served in their rights converge for the concept of sustainable development. Whereas a municipality depends on many factors to develop sustainably, that is, given its population and thinking of the future generations, we need studies that seek to analyze the situation of the municipalities in its various aspects in order to find answers or directions to removal or replacement of obstructing the process. As well as for the development to occur in one place, one needs appropriate to their stimulus policies, the government also needs favorable framework conditions for public policies of local development induction be effective. Thus, the aim of this study was to analyze whether the municipal administrations of the state of Cearà contributed, from 2009 to 2012 for the development and sustainability of their municipalities. A Municipal Management index was constructed (IGM) for 2009 and 2012 through structure indicators of the management of municipalities. Through the use of analysis techniques of Principal Components was calculated Municipal Sustainability Index (ISM), and the same variables was made by the group of municipalities in seven clusters. To see the relationship between municipal management and sustainability, we used the cumulative distribution method by quantile regression. Variables such as spending on health, spending on education and the municipal debt proved to be very important for the sustainability of cities. It was concluded that for most municipalities there was no change in sustainability over the three years, though there have been changes of management, however the values decreased, which implies that some of the characteristics of the municipalities at the beginning of his administration in 2009 may have It has been lost, such as stocks and services that are no longer running and available to the population, councils who failed to act, municipal plans that no longer exist, joints that were undone. It was noticed that the management has a direct impact on sustainability, being instrumental in the development of the municipality, in all aspects, but mainly by the variables used here to generate adequate living conditions for its residents. / O papel dos municÃpios brasileiros sofreu grandes mudanÃas nos Ãltimos vinte e seis anos. O processo de descentralizaÃÃo permitido com a ConstituiÃÃo de 1988 teve por objetivo o fortalecimento financeiro e polÃtico dos estados e, principalmente, dos municÃpios em relaÃÃo ao governo federal. Com isso, conferiu aos municÃpios maior autonomia, pois passaram a ser âentes federativosâ. Ãreas como saÃde e educaÃÃo ganharam atribuiÃÃes prÃprias dos municÃpios e se aumentam as obrigaÃÃes administrativas, consequentemente, aumentam as financeiras e os municÃpios passaram a ter que arrecadar mais para cumprÃ-las. A administraÃÃo de um municÃpio à o que o organiza e o prepara para o desenvolvimento. A necessidade de uma gestÃo organizada para que a populaÃÃo seja bem atendida em seus direitos converge, entÃo, para este conceito de desenvolvimento sustentÃvel. Considerando que um municÃpio depende de muitos fatores para se desenvolver com sustentabilidade, ou seja, atendendo sua populaÃÃo e pensando nas geraÃÃes futuras, sÃo necessÃrios estudos que procuram analisar a situaÃÃo dos municÃpios em seus mais variados aspectos para que se encontrem respostas ou direcionamentos para a remoÃÃo ou substituiÃÃo dos entraves ao processo. Assim como, para que ocorra o desenvolvimento em um local, precisa-se de polÃticas adequadas ao seu estÃmulo, o poder pÃblico tambÃm necessita de condiÃÃes estruturais favorÃveis para que as polÃticas pÃblicas sejam efetivas. Dessa forma, o objetivo geral deste estudo foi analisar se as gestÃes municipais do estado do Cearà contribuÃram, no perÃodo de 2009 a 2012, para o desenvolvimento com sustentabilidade dos seus municÃpios. Foi calculado um Ãndice de GestÃo Municipal (IGM) para 2009 e 2012 atravÃs de indicadores de estrutura da gestÃo dos municÃpios. AtravÃs do uso das tÃcnicas de AnÃlise dos Componentes Principais foi calculado o Ãndice de Sustentabilidade Municipal (ISM), e com as mesmas variÃveis fez-se o agrupamento dos municÃpios em sete clusters. Para ver as relaÃÃes entre a gestÃo municipal e a sustentabilidade, utilizou-se o mÃtodo de distribuiÃÃo cumulativa, atravÃs da RegressÃo QuantÃlica. VariÃveis como despesa com saÃde, despesa com educaÃÃo e o endividamento municipal mostraram-se muito importantes para a sustentabilidade dos municÃpios. Concluiu-se que para a maior parte dos municÃpios nÃo houve mudanÃa na sustentabilidade ao longo dos trÃs anos, porÃm houveram mudanÃas de gestÃo, entretanto os valores diminuÃram, o que pressupÃe que algumas das caracterÃsticas dos municÃpios no inÃcio de sua gestÃo em 2009 podem ter sido perdidas, como aÃÃes e serviÃos que deixaram de ser executados e oferecidos à populaÃÃo, conselhos que deixaram de atuar, planos municipais que deixaram de existir, articulaÃÃes que foram desfeitas. Percebeu-se que a gestÃo tem impacto direto sobre a sustentabilidade, sendo peÃa fundamental no desenvolvimento do municÃpio, em todos os aspectos, mas principalmente, pelas variÃveis utilizadas aqui, para gerar condiÃÃes de vida adequadas para seus habitantes.

Submitted by (edna.saturno@ufrpe.br) on 2016-06-06T14:30:31Z No. of bitstreams: 1 Micheline Correia de Souza.pdf: 1073767 bytes, checksum: b3e089ee2ef19f17792bc2ebffaac77a (MD5) / Made available in DSpace on 2016-06-06T14:30:31Z (GMT). No. of bitstreams: 1 Micheline Correia de Souza.pdf: 1073767 bytes, checksum: b3e089ee2ef19f17792bc2ebffaac77a (MD5) Previous issue date: 2013-02-25 / The present work aimed to analyze the evolution of the wage differentials among gender, after the change of the economic scenario with increasing structural investments in Pernambuco. As well as, to estimate the wage differentials by gender for different groups of work by means of a linear regression model and observe the socioeconomic profile of the workforce in Pernambuco, over the years 2000. The data used were those of PNAD, collected by IBGE, in the years 2001 and 2009. The sample matched 24,964 people in the year 2001 and 24,600 in the year 2009. To estimate the wage differentials by gender for different groups of occupation have been used the OLS and quantile regression. In General, there was a decrease in the wage differential between men and women for the estimates through the OLS, as well as for the quantile series. As well as, on an analysis by occupational groups the differential effect also decreases. Note that the categories of workers and entrepreneurs registered presented the biggest differential. It is therefore concluded that the effects of workers ' productive characteristics (education, experience, number of children) have enough correlation in the determination of wages. And to the variable of interest "genre" discrimination for women is still quite visible, but the reasons may be subjective to the local labour market in the face of the advance of structural investments and new requirements arising from the lack of skilled labor in the analysis period. / O presente trabalho objetivou analisar a evolução dos diferenciais de salários, entre gênero, após a mudança do cenário econômico com o aumento dos investimentos estruturais em Pernambuco. Assim como, estimar os diferenciais de salário por gênero para diferentes grupos de ocupação do trabalho, através de um modelo de regressão linear e observar o perfil socioeconômico da mão de obra em Pernambuco, ao longo dos anos 2000. Os dados utilizados foram os da PNAD, coletados pelo IBGE, nos anos de 2001 e 2009. A amostra correspondeu 24.964 pessoas no ano de 2001 e 24.600 no ano de 2009 em Pernambuco. Para estimar os diferenciais de salários por gênero para diferentes grupos de ocupação de trabalho foram utilizados o MQO e de regressão quantílica. De forma geral, houve uma diminuição do diferencial de salário entre o homem e a mulher para as estimações através do MQO, bem como para a série quantílica. Assim como, na análise por grupos ocupacionais o efeito do diferencial também diminui. Ressalte-se que as categorias de empresário e dos trabalhadores com carteira assinada apresentaram o maior diferencial. Conclui-se, portanto, que os efeitos das características produtivas dos trabalhadores (educação, experiência, número de filhos) apresentam bastante correlação na determinação dos salários. E para a variável de interesse “gênero” a discriminação para a mulher ainda é bastante observada, porém os motivos podem ser subjetivos ao mercado de trabalho local diante do avanço dos investimentos estruturais e das novas exigências surgidas pela falta de mão de obra especializada no período analisado.

nÃo hà / This study applies traditional techniques in Finance and Econometrics in order to analyze the impacts of Financial Crisis on some sectors of the Brazilian economy based upon market indicators provided by Getulio Vargas Foundation (FGV). Initially we apply the theory proposed by Markowitz to sectoral indicators for eight economic sectors and estimate efficient portfolios in the pre and post-financial crisis periods and we verify that the weights established in the two cases differ dramatically. After that, we estimate quantile regressions for three sectors: Mining, Metallurgic and Textiles are estimated confronting the its returns against the return of the market portfolio and the implicit volatility measured. First of all, the model captures the increase in the risk premium demanded by investors in times of crisis; in spite of the models allow us to infer that there is a change in consumer behavior in times of economic instability in order to make him more risk-tolerant. / O estudo utiliza tÃcnicas tradicionais de FinanÃas e Econometria para analisar os impactos da crise financeira de 2008 sobre alguns setores da economia brasileira, tomando por base os indicadores setoriais de mercado da FundaÃÃo GetÃlio Vargas (FGV). Inicialmente aplica-se a teoria de Markowitz aos indicadores setoriais de mercado de oito setores e estimam-se portfÃlios eficientes no perÃodo prà e pÃscrise financeira, constatando que os pesos atribuÃdos aos dois perÃodos diferem dramaticamente. Posteriormente, regressÃes quantÃlicas para os setores MineraÃÃo, Metalurgia e TÃxtil sÃo estimadas, confrontando o retorno setorial com o retorno da carteira de mercado e a volatilidade implÃcita. AlÃm de captar a elevaÃÃo do prÃmio de risco exigido pelos investidores em perÃodos de crise, os modelos permitem inferir que hà uma mudanÃa de comportamento do consumidor em perÃodos de instabilidade econÃmica no sentido de tornÃ-lo mais tolerante ao risco.

In the real world, many problems are continuous in nature. In some cases, finding the global solutions for these problems is di±cult. The reason is that the problem's objective function is non convex, nor concave and even not differentiable. Tackling these problems is often computationally too expensive. Although the development in computer technologies are increasing the speed of computations, this often is not adequate, particularly if the size of the problem's instance are large. Applying exact methods on some problems may necessitate their linearisation. Several new ideas using heuristic approaches have been considered particularly since they tackle the problems within reasonable computational time and give an approximate solution. In this thesis, the variable neighbourhood search (VNS) metaheuristic (the framework for building heuristic) has been considered. Two variants of variable neighbourhood search metaheuristic have been developed, continuous variable neighbourhood search and reformulation descent variable neighbourhood search. The GLOB-VNS software (Drazic et al., 2006) hybridises the Microsoft Visual Studio C++ solver with variable neighbourhood search metaheuristics. It has been used as a starting point for this research and then adapted and modified for problems studied in this thesis. In fact, two problems have been considered, censored quantile regression and the circle packing problem. The results of this approach for censored quantile regression outperforms other methods described in the literature, and the near-optimal solutions are obtained in short running computational time. In addition, the reformulation descent variable neighbourhood search variant in solving circle packing problems is developed and the computational results are provided.

L'objectif de cette thèse consiste à étudier les liens entre les mécanismes internes de gouvernance des banques et le comportement de prise de risque. Nous montrons qu’Adam Smith avait déjà mis en évidence la défaillance des mécanismes de gouvernance dans la Banque « Ayr » comme principal facteur induisant la prise de risque excessive et, par conséquent, sa faillite. Nous développons un modèle qui illustre qu’une rémunération indexée sur les actifs risqués n’implique pas une prise de risque plus importante. Nous constatons, aussi, que pour inciter le dirigeant à réaliser la meilleure combinaison d’actifs, le conseil d’administration est tenu de lui payer la rémunération la plus élevée. La thèse porte également sur l’étude des attributs standards du CA et ceux liés à la gouvernance des risques dans les banques de l'UE durant la période 2005-2015. Les résultats de la régression panel à effet fixe indiquent que les caractéristiques du CA affectent le niveau des crédits non performants et l’insolvabilité des banques de l’UE. Les résultats de la régression quantile à effet fixe révèlent une hétérogénéité dans la relation entre le risque bancaire et les attributs étudiés. Plus précisément, nous constatons que l’effet positif de l’indépendance et la fréquence des réunions du CA sur la gestion des risques bancaires est plus important dans les banques les plus risquées. Nos résultats mettent en évidence, également, que la prévention des comportements de prise de risque excessive des banques de l’UE nécessite l’amélioration de l’efficacité des CA à travers l’établissement des comités de risque et d’audit. / The purpose of this thesis is to study the internal mechanisms of banks’ governance and their impact on the risk-taking behavior. We show that Adam Smith had already highlighted the inadequacy of the governance’ mechanisms in “Ayr” Bank as the primary factor leading to an excessive risk-taking and, consequently, to its bankruptcy. We develop a model that aims to evaluate the extent to which governance mechanisms play a moderating role on the compensation policy and the level of risk taken by the CEO. We illustrate that a remuneration indexed on risky assets does not imply a greater risk taking. We also conclude that in order to induce the CEO to achieve the best combination of assets, the board of directors (BD) is required to pay the highest compensation. The thesis also focuses on the study of standard BD attributes as well as those related to risk’ governance in EU banks from 2005 to 2015. The empirical investigation showed that certain BD features affect the level of non-performing loan and the insolvency of EU banks. The results of the fixed-effect quantile regression reveal that the effect of the standard BD and risk’ governance attributes on risk-taking is heterogeneous. More specifically, we can note that the positive effect of the independence and frequency of board meetings on bank’ risk management is more significant in the riskier banks. In addition to this, our empirical results suggests that the prevention of excessive risk taking by EU banks requires the improvement of the effectiveness of BD through the establishment of risk an audit committees.

Les applications les plus courantes des méthodes non paramétriques concernent l’estimation d’une fonction de régression (i.e. de l’espérance conditionnelle). Cependant, il est souvent intéressant de modéliser les quantiles conditionnels, en particulier lorsque la moyenne conditionnelle ne permet pas de représenter convenablement l’impact des covariables sur la variable dépendante. De plus, ils permettent d’obtenir des graphiques plus compréhensibles de la distribution conditionnelle de la variable dépendante que ceux obtenus avec la moyenne conditionnelle. À l’origine, la « quantification » était utilisée en ingénierie du signal et de l’information. Elle permet de discrétiser un signal continu en un nombre fini de quantifieurs. En mathématique, le problème de la quantification optimale consiste à trouver la meilleure approximation d’une distribution continue d’une variable aléatoire par une loi discrète avec un nombre fixé de quantifieurs. Initialement utilisée pour des signaux univariés, la méthode a été étendue au cadre multivarié et est devenue un outil pour résoudre certains problèmes en probabilités numériques. Le but de cette thèse est d’appliquer la quantification optimale en norme Lp à l’estimation des quantiles conditionnels. Différents cas sont abordés : covariable uni- ou multidimensionnelle, variable dépendante uni- ou multivariée. La convergence des estimateurs proposés est étudiée d’un point de vue théorique. Ces estimateurs ont été implémentés et un package R, nommé QuantifQuantile, a été développé. Leur comportement numérique est évalué sur des simulations et des données réelles. / One of the most common applications of nonparametric techniques has been the estimation of a regression function (i.e. a conditional mean). However it is often of interest to model conditional quantiles, particularly when it is felt that the conditional mean is not representative of the impact of the covariates on the dependent variable. Moreover, the quantile regression function provides a much more comprehensive picture of the conditional distribution of a dependent variable than the conditional mean function. Originally, the “quantization” was used in signal and information theories since the fifties. Quantization was devoted to the discretization of a continuous signal by a finite number of “quantizers”. In mathematics, the problem of optimal quantization is to find the best approximation of the continuous distribution of a random variable by a discrete law with a fixed number of charged points. Firstly used for a one-dimensional signal, the method has then been developed in the multi-dimensional case and extensively used as a tool to solve problems arising in numerical probability. The goal of this thesis is to study how to apply optimal quantization in Lp-norm to conditional quantile estimation. Various cases are studied: one-dimensional or multidimensional covariate, univariate or multivariate dependent variable. The convergence of the proposed estimators is studied from a theoretical point of view. The proposed estimators were implemented and a R package, called QuantifQuantile, was developed. Numerical behavior of the estimators is evaluated through simulation studies and real data applications.

The thesis deals with brief introduction of the quantile regression theory. The thesis is divided into three thematic parts. In the first part the thesis deals with general introduction to the quantile regression, with theoretical aspects regarding quantile regression and with basic approaches to estimation of quantile regression parameters. The second part of the thesis focuses on general and asymptotic properties of the quantile regression. Goal of this part is to compare the quantile regression with traditional OLS regression and outline its possible application. In the third part the thesis describes statistical inference, construction of the confidence intervals and testing statistical hypotheses about quantile regression parameters. The goal of this part is to introduce traditional approach and the approach based on resampling procedures and in the end of the day perform mutual comparison of different approaches eventually propose partial modification.