In a large number of applications arising in various fields of study, time series are approximated using linear MMSE estimates. Such approximations include finite order moving average and autoregressive approximations as well as the causal Wiener filter. In this dissertation, we study two topics related to the estimation of wide sense stationary (WSS) time series using linear MMSE estimates. In the first part of this dissertation, we study the asymptotic behaviour of autoregressive (AR) and moving average (MA) approximations. Our objective is to investigate how faithfully such approximations replicate the original sequence, as the model order as well as the number of samples approach infinity. We consider two aspects: convergence of spectral density of MA and AR approximations when the covariances are known and when they are estimated. Under certain mild conditions on the spectral density and the covariance sequence, it is shown that the spectral densities of both approximations converge in L2 as the order of approximation increases. It is also shown that the spectral density of AR approximations converges at the origin under the same conditions. Under additional regularity assumptions, we show that similar results hold for approximations from empirical covariance estimates. In the second part of this dissertation, we address the problem of detecting interdependence relations within a group of time series. Ideally, in order to infer the complete interdependence structure of a complex system, dynamic behaviour of all the processes involved should be considered simultaneously. However, for large systems, use of such a method may be infeasible and computationally intensive, and pairwise estimation techniques may be used to obtain sub-optimal results. Here, we investigate the problem of determining Granger-causality in an interdependent group of jointly WSS time series by using pairwise causal Wiener filters. Analytical results are presented, along with simulations that compare the performance of a method based on finite impulse response Wiener filters to another using directed information, a tool widely used in literature. The problem is studied in the context of cyclostationary processes as well. Finally, a new technique is proposed that allows the determination of causal connections under certain sparsity conditions.

在這一篇論文裡，我們討論兩個階段的封閉式等候線網路，其中服務時間的機率分配都是Phase type分配。我們猜測服務時間的機率分配和離開時間間隔的機率分配滿足一組聯立方程組。然後，我們推導出非邊界狀態的穩定機率可以被表示成 product-form的線性組合，而每個product-form可以用聯立方程組的根來構成。利用非邊界狀態的穩定機率， 我們可以求出邊界狀態的機率。最後我們建立一個求穩定機率的演算過程。利用這個演算方法，可以簡化求穩定機率的複雜度。 / In this thesis, we are concerned with the property of a two-stage closed system in which the service times are identically of phase type. We first conjecture that the 　Laplace-Stieltjes Transforms (LST) of service time distributions may satisfy a system of equations. Then we present that the stationary probabilities on the unboundary states can be written as a linear combination of product-forms. Each component of these products can be expressed in terms of roots of the system of equations. Finally, we establish an algorithm to obtain all the stationary probabilities. The algorithm is expected to work well for relatively large customers in the system.

The UK Government has set an ambitious target to achieve 15% of final energy consumption from renewable sources by 2020. High Voltage Direct Current (HVDC) technology is an attractive solution for integrating offshore wind power farms farther from the coast. In the near future, more windfarms are likely to be connected to the UK grid using HVDC links. With the onset of this fairly new technology, new challenges are inevitable. This research is undertaken to help assist with these challenges by looking at possibilities of problems with respect to faster AC/DC interaction modes, especially, on the impact of inrush currents which occur during fault-recovery transients. In addition to that, possible mitigation strategies are also investigated. Initially, the relative merits of different transformer models are analysed with respect to inrush current transient studies. The most appropriate transformer model is selected and further validated using field measurement data. A detailed electro-magnetic-transient (EMT) model of a grid-connected MMC-HVDC system is prepared in PSCAD/EMTDC to capture the key dynamics of fault-recovery transformer inrush currents. It is shown that the transformer in an MMC system can evoke inrush currents during fault recovery, and cause transient interactions with the converter and the rest of the system, which should not be neglected. It is shown for the first time through a detailed dynamic analysis that if the current sensors of the inner-current control loops are placed at the converter-side of the transformer instead of the grid-side, the inrush currents will mainly flow from the grid and decay faster. This is suggested as a basic remedial action to protect the converter from inrush currents. Afterwards, analytical calculations of peak flux-linkage magnitude in each phase, following a voltage-sag recovery transient, are derived and verified. The effects of zero-sequence currents and fault resistance on the peak flux linkage magnitude are systematically explained. A zero-sequence-current suppression controller is also proposed. A detailed study is carried out to assess the key factors that affect the maximum peak flux-linkage and magnetisation-current magnitudes, especially with regard to fault specific factors such as fault inception angle, duration and fault-current attenuation. Subsequently, the relative merits of a prior-art inrush current mitigation strategy and its implementation challenges in a grid-connected MMC converter are analysed. It is shown that the feedforward based auxiliary flux-offset compensation scheme, as incorporated in the particular strategy, need to be modified with a feedback control technique, to alleviate the major drawbacks identified. Following that, eight different feedback based control schemes are devised, and a detailed dynamic and transient analysis is carried out to find the best control scheme. The relative merits of the identified control scheme and its implementation challenges in a MMC converter are also analysed. Finally, a detailed EMT model of an islanded MMC-HVDC system is implemented in PSCAD/EMTDC and the impacts of fault-recovery inrush currents are analysed. For that, initially, a MMC control scheme is devised in the synchronous reference frame and its controllers are systematically tuned. To obtain an improved performance, an equivalent control scheme is derived in the stationary reference frame with Proportional-Resonant controllers, and incorporated in the EMT model. Following that, two novel inrush current mitigation strategies are proposed, with the support of analytical equations, and verified.

Les inondations constituent le risque naturel prédominant dans le monde et les dégâts qu'elles causent sont les plus importants parmi les catastrophes naturelles. Un des principaux facteurs expliquant les inondations sont les précipitations extrêmes. En raison des changements climatiques, l'occurrence et l'intensité de ces dernières risquent fort probablement de s'accroître. Par conséquent, le risque d'inondation pourrait vraisemblablement s'intensifier. Les impacts de l'évolution des précipitations extrêmes sont désormais un enjeu important pour la sécurité du public et pour la pérennité des infrastructures. Les stratégies de gestion du risque d'inondation dans le climat futur sont essentiellement basées sur les simulations provenant des modèles numériques de climat. Un modèle numérique de climat procure notamment une série chronologique des précipitations pour chacun des points de grille composant son domaine spatial de simulation. Les séries chronologiques simulées peuvent être journalières ou infra-journalières et elles s'étendent sur toute la période de simulation, typiquement entre 1961 et 2100. La continuité spatiale des processus physiques simulés induit une cohérence spatiale parmi les séries chronologiques. Autrement dit, les séries chronologiques provenant de points de grille avoisinants partagent souvent des caractéristiques semblables. De façon générale, la théorie des valeurs extrêmes est appliquée à ces séries chronologiques simulées pour estimer les quantiles correspondants à un certain niveau de risque. La plupart du temps, la variance d'estimation est considérable en raison du nombre limité de précipitations extrêmes disponibles et celle-ci peut jouer un rôle déterminant dans l'élaboration des stratégies de gestion du risque. Par conséquent, un modèle statistique permettant d'estimer de façon précise les quantiles de précipitations extrêmes simulées par un modèle numérique de climat a été développé dans cette thèse. Le modèle développé est spécialement adapté aux données générées par un modèle de climat. En particulier, il exploite l'information contenue dans les séries journalières continues pour améliorer l'estimation des quantiles non stationnaires et ce, sans effectuer d'hypothèse contraignante sur la nature de la non-stationnarité. Le modèle exploite également l'information contenue dans la cohérence spatiale des précipitations extrêmes. Celle-ci est modélisée par un modèle hiérarchique bayésien où les lois a priori des paramètres sont des processus spatiaux, en l'occurrence des champs de Markov gaussiens. L'application du modèle développé à une simulation générée par le Modèle régional canadien du climat a permis de réduire considérablement la variance d'estimation des quantiles en Amérique du Nord. / Precipitation extremes plays a major role in flooding events and their occurrence as well as their intensity are expected to increase. It is therefore important to anticipate the impacts of such an increase to ensure the public safety and the infrastructure sustainability. Since climate models are the only tools for providing quantitative projections of precipitation, flood risk management for the future climate may be based on their simulations. Most of the time, the Extreme value theory is used to estimate the extreme precipitations from a climate simulation, such as the T-year return levels. The variance of the estimations are generally large notably because the sample size of the maxima series are short. Such variance could have a significant impact for flood risk management. It is therefore relevant to reduce the estimation variance of simulated return levels. For this purpose, the aim of this paper is to develop a non-stationary and regional statistical model especially suited for climate models that estimates precipitation extremes. At first, the non-stationarity is removed by a preprocessing approach. Thereafter, the spatial correlation is modeled by a Bayesian hierarchical model including an intrinsic Gaussian Markov random field. The model has been used to estimate the 100-year return levels over North America from a simulation by the Canadian Regional Climate Model. The results show a large estimation variance reduction when using the regional model.

Simplifications regarding calculations of fatigue life due to vibrational loading is based on weakly stationarity assumptions which is a time independent method. The hypothesis was based on the uncertainty of these assumptions. The aim of this study was to examine whether the analysed data fulfilled the assumptions of weakly stationarity. It was determined that the assumption was not valid for most signals and a comparison of time dependent methods should be performed to evaluate the difference compared with the time independent method. Two time dependent methods were constructed and implemented on the signals based on the results of performed stationarity tests. The result determined that a decrease in fatigue life of an investigated weld might occur for the two time dependent methods compared with the time independent method. The method which was considered to produce the most accurate results was also the most constrained as to the amount of data that fulfilled its requirements. A conclusion was drawn that signals containing more data was necessary to achieve conclusive results of the fatigue life. The hypothesis was proven to be mostly true since most of the analysed signals were found to be piecewise weakly stationary.

Soit μ une mesure de probabilité borélienne sur SL m+1 (R) tel que le sous-groupe engendré par le support de μ est Zariski dense. Soit V une représentation irréductible de dimension finie de SL m+1 (R). D’après un théorème de Furstenberg, il existe une unique mesure μ-stationnaire sur PV et nous nous somme intéressés à la décroissance de Fourier de cette mesure. Le résultat principal de cette thèse est que la transformée de Fourier de la mesure stationnaire a une décroissance polynomiale. À partir de ce résultat, nous obtenons un trou spectral de l’opérateur de transfert, dont les propriétés nous permettent d’établir un terme d’erreur exponentiel pour le théorème de renouvellement dans le cadre des produits de matrices aléatoires. L’ingrédient essentiel est une propriété de décroissance de Fourier des convolutions multiplicatives de mesures sur R n , qui est une généralisation d’un théorème de Bourgain en dimension 1. Nous établissons cet ingrédient en utilisant un estimée somme produit de He et de Saxcé.Dans la dernière partie, nous généralisons un résultat de Lax et Phillips et un résultat de Hamenstädt sur la finitude des petites valeurs propres de l’opérateur de Laplace sur les variétés hyperboliques géométriquement finies. / Let μ be a Borel probability measure on SL m+1 (R), whose support generates a Zariski dense subgroup. Let V be a finite dimensional irreducible linear representation of SL m+1 (R). A theorem of Furstenberg says that there exists a unique μ-stationary probability measure on PV and we are interested in the Fourier decay of the stationary measure. The main result of the thesis is that the Fourier transform of the stationary measure has a power decay. From this result, we obtain a spectral gap of the transfer operator, whose properties allow us to establish an exponential error term for the renewal theorem in the context of products of random matrices. A key technical ingredient for the proof is a Fourier decay of multiplicative convolutions of measures on R n , which is a generalisation of Bourgain’s theorem on dimension 1. We establish this result by using a sum-product estimate due to He-de Saxcé. In the last part, we generalize a result of Lax-Phillips and a result of Hamenstädt on the finiteness of small eigenvalues of the Laplace operator on geometrically finite hyperbolic manifolds

O objetivo principal deste trabalho é o desenvolvimento de controladores robustos baseados no princípio do modelo interno, em referenciais síncrono e estacionário, para aplicação ao estágio de saída de uma fonte ininterrupta de energia (UPS) a fim de minimizar a distorção na tensão de saída causada pela conexão de cargas não lineares balanceadas e desbalanceadas. A formulação em referencial estacionário (abc) é realizada através da aplicação de controladores com múltiplos modos ressonantes, a fim de se estabelecer erro nulo ao seguimento de referência senoidal e rejeição de distúrbios na tensão de saída devido às correntes com elevado conteúdo harmônico drenadas pelas cargas. Além disso, o controle é formulado em referencial síncrono (dq0) utilizando controladores Proporcional-Integral (PI) convencionais muito difundidos na maioria das aplicações comerciais de UPS. O projeto de ambos controladores é realizado utilizando uma metodologia de controle robusto com realimentação de estados, onde os parâmetros dos controladores são determinados através da resolução de um problema de otimização convexa sujeito a um conjunto de restrições na forma de desigualdades matriciais lineares (LMI). Uma análise comparativa de desempenho é realizada entre controladores com um modo ressonante (sintonizado na fundamental) e o PI em dq0, pois apresentam estruturas funcionalmente equivalentes sob a ótica do princípio do modelo interno aplicada a seus respectivos referenciais. Além do mais, demonstra-se a melhoria no desempenho com o uso dos controladores múltiplo ressonantes em referencial estacionário onde escolhe-se as frequências de ressonância de cada modo de maneira a suprimir os efeitos de harmônicas específicas na tensão de saída da UPS. A análise comparativa entre os controladores propostos é realizada através de simulações numéricas, utilizando os procedimentos de ensaio dinâmico e estático e as exigências estabelecidas pela norma internacional IEC 62040-3. / The main objective of this work is the development of robust controllers based on the internalmodel principle, in synchronous and stationary frames, applied to the output stage of an uninterruptible power supply (UPS), in order to minimize the output voltage distortion caused by the connection of balanced and unbalanced nonlinear loads. The formulation in stationary abc-frame is accomplished through the aplication of a multiple resonant controller, so that, it is possible to achieve zero-error tracking of the sinusoidal reference and disturbances rejection on the output voltage due to the high amount of harmonic currents drained by the loads. Moreover, a controller in synchronous reference frame (dq0 axis) is formulated through the application of conventional Proportional-Integral (PI) controllers which are widely used in comercial UPS applications. The design of both controllers is formulated using a state-feedback robust controlmethod, in which the controller parameters are determined by solving a convex optimization problem subject to a set of LMI constraints. A comparative analysis on the performance of the single-mode resonant controller (tuned at the fundamental frequency) and the PI controller is performed, because these controllers are functionally equivalent in the sense of the internal model principle applied to their respective frames. Furthermore, the improvement in performance is demostrated with the use of multiple resonant controllers in stationary abc-frame where the resonance frequencies are chosen to suppress the effects of a specific harmonic in the UPS output voltage. The comparative analysis of the proposed controllers is performed through numerical simulations, making use of the dynamical and steady-state test methods and performance requirements defined by the IEC 62040-3 international stardard.

Neste trabalho estudamos uma dinâmica estocástica de partículas de duas espécies baseada em células. Basicamente, incorporamos algumas inovações em um modelo unidimensional proposto e resolvido por R. da Silva et al. (Physica A, 2015), que considera que em um célula, na ausência de partículas da espécie contrária, a partícula vai pra frente com uma probabilidade p, que representaria um campo na direção longitudinal de um corredor e fica na própria célula com q=1-p. Contudo, essa probabilidade p é reduzida de acordo com a concentração de partículas contrárias. Nosso trabalho não apenas estendeu o problema pra duas dimensões como também incluiu aspectos relativos a colisão e o espalhamento para células vizinhas. Nossos resultados são divididos em duas situações: a) Espécie contrária permanece imóvel funcionando como obstáculos b) Espécie contrária em movimento. Na primeira situação podemos ver uma interessante transição na distribuição dos tempos de travessia em função das concentrações dos obstáculos, por monitorar a curtose da distribuição. Quando a espécie contrária se movimenta, vemos que o tempo de destilação entre as partículas (tempo para que as espécies estejam geograficamente separadas no corredor) depende do parâmetro ligado ao espalhamento transversal das partículas, parâmetro este, que não influencia no caso das partículas paradas. Finalmente nós colocamos as partículas em um sistema com condições periódicas de contorno. Neste caso, podemos observar o aparecimento de padrões de bandas longitudinais ao campo, exatamente como ocorrem em problemas de coloides carregados sob a ação de campos longitudinais e em modelos de pedestres em corredores. Mostramos como o sistema relaxa para tal tipo de estado estacionário utilizando um adequado parâmetro de ordem ligado a segregação das partículas. Nosso modelo, diferentemente dos modelos para pedestres, não se baseia em equações tipo Langevin. Nossa abordagem é totalmente estocástica e por esse ponto de vista ainda mais fundamental e geral, podendo ser estendida para mais modelos de partículas em fluxos contrários. Nossa solução vem tanto através de simulações Monte Carlo bem como soluções das equações diferenciais parciais que descrevem o sistema e que são oriundas das recorrências estabelecidas para os caminhantes aleatórios. As simulações Monte Carlo e soluções via EDP mostram boa concordância em todos os aspectos analisados, tanto qualitativa quanto quantitativamente. / In this work we study a stochastic dynamic of particles of two types based on cells. Basically we incorporate some innovations on a one-dimensional model proposed and solved by R. da Silva et al. (Physica A, 2015) which considers that in the absence of particles of the opposite species in the cell a particle goes toward the next cell with probability p and returns to the previous cell with probability q = 1 p. However this motion probability linearly decreases with the relative density of the contrary species. Our work not only expands the problem for two dimensions but also includes collision aspects by adding scattering to the neighbouring cells. Our results are divided into two di erent categories: a) One of the species remain xed in their places which means that such particles will work as obstacles; b) Both species can move in the environment. In the rst situation we can observe, by monitoring the kurtosis, that an interesting transition of the crossing time distribution arises as the concentration of the obstacles increases. When both species can move we can observe that the distillation time (spent time for the complete geographical separation of the species in the corridor) depends on the parameter related to the perpendicular scattering of the particles. This same parameter has shown no in uence over the time distributions in the rst situation. Finally we implement periodic boundary conditions in the eld's direction. In this case we are able to observe the arising of band patterns parallel to the eld's direction exactly as it does with oppositely charged colloids under the in uence of a uniform electric eld or pedestrian dynamics in corridors. We also show how the system relax to such stationary state by using a suitable order parameter related to the particles segregation. Di erently from other pedestrian dynamics models, our model is not based on a Langevin-type equation. Our approach is totally stochastic and from this point of view, more fundamental and general to be extended to more types of models considering particles under counter ow. Our solution is obtained by both Monte Carlo simulations and numerical integration of partial di erential equations (PDE) from recurrence relation of the directed random walkers. The Monte Carlo simulations and the solutions of the PDE show a good agreement in all aspects analysed both qualitatively and quantitatively.

Sistemas cujos componentes interagem por meio de forças de longo alcance não-blindadas por exemplo, sistemas estelares e plasmas não-neutros têm algumas características anô- malas em relação a sistemas com forças blindadas ou de curto alcance. Além de apresentarem características termodinâmicas peculiares como calor especí co negativo e inequivalência de ensembles, sua dinâmica é predominantemente não-colisional e leva à estados quasiestacion ários fora de equilíbrio. Esses estados são notoriamente difíceis de prever dada uma condição inicial qualquer, e ainda não existe uma teoria uni cada para tratá-los. O equilíbrio termodinâmico é atingido somente após tempos longos que escalam com o tamanho do sistema, muitas vezes excedendo o tempo de vida do universo. A relaxação para o equilíbrio, portanto, tem duas escalas de tempo: uma, curta, que leva a estados quasi-estacionários fora de equilíbrio, e a segunda, longa, que leva ao equilíbrio termodinâmico. Nesta tese de doutorado, examinamos esses fenômenos aplicando modelos teóricos e simulação numérica para diferentes sistemas de interação de longo-alcance, incluindo um modelo de spins clássicos tipo XY com longo alcance, e o sistema auto-gravitante em três dimensões. Em uma segunda etapa, estudamos a relaxação para o equilíbrio termodinâmico, a relaxação colisional, através de equações cinéticas e simulação numérica. Desta forma, buscamos esclarecer os mecanismos por trás dos estados quasi-estacionários e da relaxação colisional. / Systems whose components interact by unscreened long-range forces for example, stellar systems and non-neutral plasmas have characteristics that are anomalous with respect to systems with shielded or short-range forces. Besides presenting unique thermodynamic properties such as negative speci c heat and inequivalence of ensembles, their dynamics is predominantly collisionless and leads to out-of-equilibrium quasi-stationary states. These states are notoriously di cult to predict given an arbitrary initial condition, and there is still no uni ed theory to treat them. Thermodynamic equilibrium is reached only after long timescales that increase with the system size and often exceed the lifetime of the universe. Relaxation to equilibrium, therefore, has two timescales: one short, leading to outof- equilibrium quasi-stationary states, and a second, longer, which leads to thermodynamic equilibrium. In this thesis, we examine these phenomena by applying theoretical models and numerical simulation for di erent long-range interacting systems, including a model of classical XY-type spins with long-range interactions, and the self-gravitating system in three dimensions. In a second stage we study the collisional relaxation to thermodynamic equilibrium through kinetic equations and numerical simulation. We thus seek to clarify the mechanisms behind the quasi-stationary states and collisional relaxation.

Nesta tese discutimos abordagens que utilizam cópulas para descrever dependências entre instrumentos nanceiros e avaliamos a performance destes métodos. Muitas crises nanceiras aconteceram desde o nal da década de 90, incluindo a crise asiática (1997), a crise da dívida da Rússia (1998), a crise da bolha da internet (2000), as crises após o 9/11 (2001) e a guerra do Iraque (2003), a crise do subprime or crise nanceira global (2007-08), e a crise da dívida soberana europeia (2009). Todas estas crises levaram a uma perda maciça de riqueza nanceira e a um aumento da volatilidade observada, e enfatizaram a importância de uma política macroprudencial mais robusta. Em outras palavras, perturbações nanceiras tornam os processos econômicos altamente não-lineares, levando os principais bancos centrais a tomarem medidas contrárias para conter a angústia - nanceira. Devido aos complexos padrões de dependência dos mercados nanceiros, uma abordagem multivariada em grandes dimensões para a análise da dependência caudal é seguramente mais perspicaz do que assumir retornos com distribuição normal multivariada. Dada a sua exibilidade, as cópulas são capazes de modelar melhor as regularidades empiricamente veri cadas que são normalmente atribuídas a retornos nanceiros multivariados: (1) volatilidade condicional assimétrica com maior volatilidade para grandes retornos negativos e menor volatilidade para retornos positivos (HAFNER, 1998); (2) assimetria condicional (AIT-SAHALIA; BRANDT, 2001; CHEN; HONG; STEIN, 2001; PATTON, 2001); (3) excesso de curtose (TAUCHEN, 2001; ANDREOU; PITTIS; SPANOS, 2001); e (4) dependência temporal não linear (CONT, 2001; CAMPBELL; LO; MACKINLAY, 1997). A principal contribuição dos ensaios é avaliar se abordagens mais so sticadas do que o método da distância e o tradicional modelo de Markowitz podem tirar proveito de quaisquer anomalias/fricções de mercado. Os ensaios são uma tentativa de fornecer uma análise adequada destas questões usando conjuntos de dados abrangentes e de longo prazo. Empiricamente, demonstramos que as abordagens baseadas em cópulas são úteis em todos os ensaios, mostrando-se bené cas para modelar dependências em diferentes cenários, avaliando as medidas de risco caudais mais adequadamente e gerando rentabilidade superior a dos benchmarks utilizados. / In this thesis we discuss copula-based approaches to describe statistical dependencies within nancial instruments and evaluate its performance. Many nancial crises have occurred since the late 1990s, including the Asian crisis (1997), the Russian national debt crisis (1998), the dot-com bubble crisis (2000), the crises after 9-11 (2001) and Iraq war (2003), the subprime mortgage crisis or global nancial crisis (2007-08), and the European sovereign debt crisis (2009). All of these crises lead to a massive loss of nancial wealth and an upward in observed volatility and have emphasized the importance of a more robust macro-prudential policy. In other words, nancial disruptions make the economic processes highly nonlinear making the major central banks to take counter-measures in order to contain nancial distress. The methods for modeling uncertainty and evaluating the market risk on nancial markets are now under more scrutiny after the global nancial crisis. Due to the complex dependence patterns of nancial markets, a high-dimensional multivariate approach to tail dependence analysis is surely more insightful than assuming multivariate normal returns. Given its exibility, copulas are able to model better the empirically veri ed regularities normally attributed to multivariate nancial returns: (1) asymmetric conditional volatility with higher volatility for large negative returns and smaller volatility for positive returns (HAFNER, 1998); (2) conditional skewness (AITSAHALIA; BRANDT, 2001; CHEN; HONG; STEIN, 2001; PATTON, 2001); (3) excess kurtosis (TAUCHEN, 2001; ANDREOU; PITTIS; SPANOS, 2001); and (4) nonlinear temporal dependence (CONT, 2001; CAMPBELL; LO; MACKINLAY, 1997). The principal contribution of the essays is to assess if more sophisticated approaches than the distance method and plain Markowitz model can take advantage of any market anomalies/ fricctions. The essays are one attempt to provide a proper analysis in these issues using a long-term and comprehensive datasets. We empirically show that copula-based approaches are useful in all essays, proving bene cial to model dependencies in di erent scenarios, assessing the downside risk measures more adequately and yielding higher profitability than the benchmarks.