Portfolio Management is the process of allocating an investor's wealth to in­ vestment opportunities over a given planning period. Not only should Portfolio Management be treated within a multi-period framework, but one should also take into consideration the stochastic nature of related parameters. After a short review of key concepts from Finance Theory, e.g. utility function, risk attitude, Value-at-rusk estimation methods, a.nd mean-variance efficiency, this work describes a framework for the formulation of the Portfolio Management problem in a Stochastic Programming setting. Classical solution techniques for the resolution of the resulting Stochastic Programs (e.g. L-shaped Decompo­ sition, Approximation of the probability function) are presented. These are discussed within both the two-stage and the multi-stage case with a special em­ phasis on the former. A description of how Importance Sampling and EVPI are used to improve the efficiency of classical methods is presented. Postoptimality Analysis, a sensitivity analysis method, is also described. / Statistics / M. Sc. (Operations Research)

Light transport simulation is used in realistic image synthesis to create physically plausible images of virtual scenes. Important components of the scenes are participating media (e.g. air, water, skin etc.). Efficient computation of light transport in participating media robust to their large diversity is still an open problem. We implemented the UPBP algorithm recently developed by Křivánek et al. It addresses the problem by combining several complementary previous methods using multiple importance sampling, and excels at rendering scenes where the previous methods alone fail. The implementation is available online, we focused on its thorough description to facilitate and support further research in this field. Powered by TCPDF (www.tcpdf.org)

Path tracing is a basic, statistically unbiased method for calculating the global illumination in 3D scenes. For practical purposes, the algorithm is too slow, so it is used mainly for theoretical purposes or as a base for more advanced algorithms. This thesis explores the possibility of improving this algorithm by augmenting the sampling part, which computes outgoing directions during ray traversal through the scene. This optimization is accomplished by creating a special data structure in a preprocess step, which describes approximate light distribution in the scene and which then aids the sampling process. The presented algorithm is implemented in the PBRT library.

[pt] Esta dissertação discute o modelo de Volatilidade Estocástica (SV) estimado via metodologia Durbin & Koopman, chamada Verossimilhança de Monte Carlo( MCL). Comparou-se a cobertura condicional do valor em risco (VaR), deste modelo, com as do modelo GARCH(1,1) e SV estimado via Quasi Máxima Verossimilhança (QML). Os modelos foram estendindos a distúrbios Gaussiano e t-Student na equação da média. O desempenho dos modelos foi avaliado fora da amostra para retornos diários dos índices Ibovespa, S&P500, Nasdaq e Dow Jones. Para o critério de avaliação foi utilizado o teste de Christoffersen. Foram econtradas evidências empíricas de que o modelo SV estimado via MCL é tão eficiente quanto o modelo GARCH(1,1), em termos da cobertura condicional do VaR. / [en] This dissertation discusses the estimation of the Stochastic Volatility (SV)model using a Durbin and Koopman methodology called Monte Carlo Like-lihood (MCL). The conditional coverage of value at risk (VaR) of SV via MCL model was compared to the GARCH (1,1) model and to the SV model via Quasi Maximum Likelihood (QML) estimation. The models were extended to Gaussian and Student-t isturbances in the mean equation. The performances of the models were evaluated out-of-sample for daily returns on the Ibovespa, S&P500, Nasdaq and Dow Jones indexes. Christoffersen test were applied for the evaluation criteria. In terms of the VaR conditional coverage, empirical evidences indicate that the SV model via MCL estimation is as efficient as the GARCH (1,1) model.

Neste trabalho definimos o processo de exclusão simples simétrico em tempo discreto sobre grafos por meio de permutações com restrições sobre os índices dos vértices dos grafos. O processo é uma generalização das permutações dos índices do grafo completo. Apresentamos algoritmos de Monte Carlo e de amostragem sequencial por importância para amostrar permutações com restrições inspirados pelo problema análogo de calcular permanentes. Como aplicação, utilizamos esses algoritmos para estimar os tempos de relaxação do processo de exclusão simples simétrico em tempo discreto sobre grafos aleatórios densos de Erdös-Rényi com laços / In this work we define the symmetric simple exclusion process in discrete time over graphs by means of suitably restricted permutations over the labels of the vertices of the graphs. The process is a generalization of the shuffling of labels on the complete graph. Straightforward Monte Carlo and sequential importance sampling algorithms to sample restricted permutations inspired by the related problem of computing permanents are discussed. We illustrate the formalism by estimating the relaxation times of the symmetric simple exclusion process in discrete time over dense loop-augmented Erdös-Rényi random graphs

Ce travail se décompose en deux grandes parties, "Estimations répétées dans le cadre de la modélisation bayésienne" et "Modélisation de la transmission de maladies infectieuses dans une population. Estimation des paramètres.". Les techniques développées dans la première partie sont utilisées en fin de la seconde partie. La première partie est consacrée à des optimisations d'algorithmes stochastiques très souvent utilisés, notamment dans le contexte des modélisations Bayésiennes. Cette optimisation est particulièrement faite lors de l'étude empirique d'estimateurs des paramètres d'un modèle où les qualités des estimateurs sont évaluées sur un grand nombre de jeux de données simulées. Quand les lois a posteriori ne sont pas explicites, le recours à des algorithmes stochastiques itératifs (de la famille des algorithmes dits de Monte Carlo par Chaîne de Makov) pour approcher les lois a posteriori est alors très couteux en temps car doit être fait pour chaque jeu de données. Dans ce contexte, ce travail consiste en l'étude de solutions évitant un trop grand nombre d'appels à ces algorithmes mais permettant bien-sûr d'obtenir malgré tout des résultats précis. La principale technique étudiée dans cette partie est celle de l'échantillonnage préférentiel. La seconde partie est consacrée aux études de modèles épidémiques, en particulier le modèle compartimental dit SIS (Susceptible-Infecté-Susceptible) dans sa version stochastique. L'approche stochastique permet de prendre en compte l'hétérogénéité de l'évolution de la maladie dans la population. les approches par des processus Markoviens sont étudiés où la forme des probabilités de passage entre les états est non linéaire. La solution de l'équation différentielle en probabilité n'est alors en général pas explicite. Les principales techniques utilisées dans cette partie sont celles dites de développement de l'équation maîtresse ("master equation") appliquées au modèle SIS avec une taille de population constante. Les propriétés des estimateurs des paramètres sont étudiées dans le cadre fréquentiste et bayésien. Concernant l'approche Bayésienne, les solutions d'optimisation algorithmique de la première partie sont appliquées. / This work consists in two parts, "Repeated estimates in bayesian modelling " and " Modelling of the transmission of infectious diseases in a population. Estimation of the parameters". Techniques developed in the first part are used at the end of the second part.The first part deals with optimizations of very often used stochastic algorithms, in particular in the context of Bayesian modelling. This optimization is particularly made when empirical study of estimates based on numerous simulated data sets is done. When posterior distribution of parameters are not explicit, its approximation is obtained via iterative stochastic algorithms (of the family of Markov Chain Monte Carlo) which is computationally expensive because has to be done on each data set. In this context, solutions are proposed avoiding an excess large number of MCMC calls but nevertheless giving accurate results. The Importance Sampling method is used in combination with MCMC in Bayesian simulation study. The second part deals with epidemic models, in particular the compartimental model SIS (Susceptible-Infectious-Susceptible) in its stochastic version. The stochastic approach allows to take into account the heterogeneousness of disease evolution in the population. Markov Process is particularly studied where transition probability between states is not linear, the solution of the differential equation in probability being then generally not explicit. The main techniques used in this part are the ones based on Master equation applied on SIS model with a constant population size. Empirical properties of parameters estimates are studied in frequentist and Bayesian context with algorithmic optimization presented in the first part.

Les travaux de recherche constituent une contribution au développement de la théorie de la fiabilité en mécanique des fluides. Pour la conception de machines et de systèmes mécatroniques complexes, de nombreux composants fluides, difficiles à dimensionner, sont utilisés. Ces derniers ont des caractéristiques intrinsèques statiques et dynamiques sensibles et ont donc une grande importance sur la fiabilité et la durée de vie de la plupart des machines et des systèmes.Le développement effectué se concentre spécialement sur l'évaluation de la fiabilité d’un palier fluide grâce à un couplage « mécanique des fluides - fiabilité ». Ce couplage exige une définition propre de la fonction d’état limite permettant d’estimer la probabilité de défaillance d’un palier fluide. La modélisation par l'équation de Reynolds modifiée permet de déterminer la capacité de charge d’un palier fluide en fonction des conditions de fonctionnement. Plusieurs formes simples de paliers fluides ont été modélisées analytiquement et leurs probabilités de défaillance ont été estimées grâce à des méthodes d'approximation FORM/SORM (First Order Reliability, Second Order Reliability) et de simulation Monte Carlo. / These research is a contribution to the development of reliability theory in fluid mechanics. For the machines design and complex mechatronic systems, many fluid components are used. These components have static and dynamic sensitive characteristics and thus have a great significance on the reliabilityand lifetime of the machines and systems. Development performed focuses specifically on the reliability evaluation of a fluid bearing using a"fluid mechanics - reliability" interaction approach. This coupling requires a specific definition of the limit state function for estimating the failure probability of a fluid bearing. The Reynolds equation permits to determine the fluid bearing load capacity according to the operating conditions. Several simple geometries of fluid bearings were modeled analytically and their failure probabilities were estimated using the approximation methods FORM / SORM (First Order Reliability Method,Second Order Reliability Method) and Monte Carlo simulation.

Neste trabalho definimos o processo de exclusão simples simétrico em tempo discreto sobre grafos por meio de permutações com restrições sobre os índices dos vértices dos grafos. O processo é uma generalização das permutações dos índices do grafo completo. Apresentamos algoritmos de Monte Carlo e de amostragem sequencial por importância para amostrar permutações com restrições inspirados pelo problema análogo de calcular permanentes. Como aplicação, utilizamos esses algoritmos para estimar os tempos de relaxação do processo de exclusão simples simétrico em tempo discreto sobre grafos aleatórios densos de Erdös-Rényi com laços / In this work we define the symmetric simple exclusion process in discrete time over graphs by means of suitably restricted permutations over the labels of the vertices of the graphs. The process is a generalization of the shuffling of labels on the complete graph. Straightforward Monte Carlo and sequential importance sampling algorithms to sample restricted permutations inspired by the related problem of computing permanents are discussed. We illustrate the formalism by estimating the relaxation times of the symmetric simple exclusion process in discrete time over dense loop-augmented Erdös-Rényi random graphs

Financial time series studies indicate that the lognormal assumption for the return of an underlying security is often violated in practice. This is due to the presence of time-varying volatility in the return series. The most common departures are due to a fat left-tail of the return distribution, volatility clustering or persistence, and asymmetry of the volatility. To account for these characteristics of time-varying volatility, many volatility models have been proposed and studied in the financial time series literature. Two main conditional-variance model specifications are the autoregressive conditional heteroscedasticity (ARCH) and the stochastic volatility (SV) models. The SV model, proposed by Taylor (1986), is a useful alternative to the ARCH family (Engle (1982)). It incorporates time-dependency of the volatility through a latent process, which is an autoregressive model of order 1 (AR(1)), and successfully accounts for the stylized facts of the return series implied by the characteristics of time-varying volatility. In this thesis, we review both ARCH and SV models but focus on the SV model and its variations. We consider two modified SV models. One is an autoregressive process with stochastic volatility errors (AR--SV) and the other is the Markov regime switching stochastic volatility (MSSV) model. The AR--SV model consists of two AR processes. The conditional mean process is an AR(p) model , and the conditional variance process is an AR(1) model. One notable advantage of the AR--SV model is that it better captures volatility persistence by considering the AR structure in the conditional mean process. The MSSV model consists of the SV model and a discrete Markov process. In this model, the volatility can switch from a low level to a high level at random points in time, and this feature better captures the volatility movement. We study the moment properties and the likelihood functions associated with these models. In spite of the simple structure of the SV models, it is not easy to estimate parameters by conventional estimation methods such as maximum likelihood estimation (MLE) or the Bayesian method because of the presence of the latent log-variance process. Of the various estimation methods proposed in the SV model literature, we consider the simulated maximum likelihood (SML) method with the efficient importance sampling (EIS) technique, one of the most efficient estimation methods for SV models. In particular, the EIS technique is applied in the SML to reduce the MC sampling error. It increases the accuracy of the estimates by determining an importance function with a conditional density function of the latent log variance at time t given the latent log variance and the return at time t-1. Initially we perform an empirical study to compare the estimation of the SV model using the SML method with EIS and the Markov chain Monte Carlo (MCMC) method with Gibbs sampling. We conclude that SML has a slight edge over MCMC. We then introduce the SML approach in the AR--SV models and study the performance of the estimation method through simulation studies and real-data analysis. In the analysis, we use the AIC and BIC criteria to determine the order of the AR process and perform model diagnostics for the goodness of fit. In addition, we introduce the MSSV models and extend the SML approach with EIS to estimate this new model. Simulation studies and empirical studies with several return series indicate that this model is reasonable when there is a possibility of volatility switching at random time points. Based on our analysis, the modified SV, AR--SV, and MSSV models capture the stylized facts of financial return series reasonably well, and the SML estimation method with the EIS technique works very well in the models and the cases considered.

Financial time series studies indicate that the lognormal assumption for the return of an underlying security is often violated in practice. This is due to the presence of time-varying volatility in the return series. The most common departures are due to a fat left-tail of the return distribution, volatility clustering or persistence, and asymmetry of the volatility. To account for these characteristics of time-varying volatility, many volatility models have been proposed and studied in the financial time series literature. Two main conditional-variance model specifications are the autoregressive conditional heteroscedasticity (ARCH) and the stochastic volatility (SV) models. The SV model, proposed by Taylor (1986), is a useful alternative to the ARCH family (Engle (1982)). It incorporates time-dependency of the volatility through a latent process, which is an autoregressive model of order 1 (AR(1)), and successfully accounts for the stylized facts of the return series implied by the characteristics of time-varying volatility. In this thesis, we review both ARCH and SV models but focus on the SV model and its variations. We consider two modified SV models. One is an autoregressive process with stochastic volatility errors (AR--SV) and the other is the Markov regime switching stochastic volatility (MSSV) model. The AR--SV model consists of two AR processes. The conditional mean process is an AR(p) model , and the conditional variance process is an AR(1) model. One notable advantage of the AR--SV model is that it better captures volatility persistence by considering the AR structure in the conditional mean process. The MSSV model consists of the SV model and a discrete Markov process. In this model, the volatility can switch from a low level to a high level at random points in time, and this feature better captures the volatility movement. We study the moment properties and the likelihood functions associated with these models. In spite of the simple structure of the SV models, it is not easy to estimate parameters by conventional estimation methods such as maximum likelihood estimation (MLE) or the Bayesian method because of the presence of the latent log-variance process. Of the various estimation methods proposed in the SV model literature, we consider the simulated maximum likelihood (SML) method with the efficient importance sampling (EIS) technique, one of the most efficient estimation methods for SV models. In particular, the EIS technique is applied in the SML to reduce the MC sampling error. It increases the accuracy of the estimates by determining an importance function with a conditional density function of the latent log variance at time t given the latent log variance and the return at time t-1. Initially we perform an empirical study to compare the estimation of the SV model using the SML method with EIS and the Markov chain Monte Carlo (MCMC) method with Gibbs sampling. We conclude that SML has a slight edge over MCMC. We then introduce the SML approach in the AR--SV models and study the performance of the estimation method through simulation studies and real-data analysis. In the analysis, we use the AIC and BIC criteria to determine the order of the AR process and perform model diagnostics for the goodness of fit. In addition, we introduce the MSSV models and extend the SML approach with EIS to estimate this new model. Simulation studies and empirical studies with several return series indicate that this model is reasonable when there is a possibility of volatility switching at random time points. Based on our analysis, the modified SV, AR--SV, and MSSV models capture the stylized facts of financial return series reasonably well, and the SML estimation method with the EIS technique works very well in the models and the cases considered.