O objetivo desta dissertação é fornecer primeiramente o arcabouço necessário para o entendimento do derivativo opções, muito utilizado nos mercados financeiros mundiais, e posteriormente executar precificações de opções americanas a partir dos modelos dos mínimos quadrados de Monte Carlo (LSM), o modelo de árvore binomial com extrapolação de Richardson e a aproximação analítica de Bjerksund e Stensland (B&S), aplicando duas plataformas de processamento paralelo computacional, a TPL (Task Parallel Library) nativa no .NET framework 4.5 e a plataforma CUDA (Compute Unified Device Architecture), demonstrando o comparativo dos resultados obtidos a cada modelo diante de cada plataforma. / The objective of this dissertation is to provide first the necessary framework for the understanding of the derivative options, widely used in the world financial markets, and later to execute the American option pricing from Monte Carlo least squares models (LSM), the binomial tree model with Richardson extrapolation and the Bjerksund and Stensland analytic approach (BJS) by applying two parallel computational processing platforms, the native TPL (Task Parallel Library) in the .NET framework 4.5 and the CUDA platform (Compute Unified Device Architecture), demonstrating the comparison of the obtained results to each model before each platform.

Orientador: Yuzo Iano / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-20T16:16:02Z (GMT). No. of bitstreams: 1 SilvestreBezerra_ManoelIvanildo_D.pdf: 5165235 bytes, checksum: 93259721102d21ca5d681ec6df5622e0 (MD5) Previous issue date: 2012 / Resumo: Neste trabalho é proposto um novo método de estimação separada (sub-ótimo) para o processo (modelo) espectral ARMA. Os métodos sub-ótimos utilizam-se das equações de Yule-Walker e do método de mínimos quadrados para as estimativas AR, e geralmente do método de Durbin para as estimativas MA. Dado que os parâmetros AR e MA já foram estimados, no método proposto é feita uma nova filtragem AR do sinal de interesse utilizando-se as estimativas da parte MA. A partir deste novo sinal estimado, determinam-se as novas estimativas das partes AR e MA do processo ARMA, e em seguida obtém-se a estimativa da densidade espectral de potência. Os resultados dependem muito do espectro de interesse, e da parametrização que foi utilizada, mas de um modo geral os resultados fornecidos foram muito bons. Um estudo descrevendo os principais métodos de estimação espectral paramétrica dos processos ARMA também é realizado neste trabalho. Esses métodos são comparados medindo a precisão através do erro relativo e do coeficiente de variação médio das estimativas dos parâmetros / Abstract: This work proposes a new method of estimating separate (sub-optimal) for the spectrum ARMA process (model). The sub-optimal methods use the Yule-Walker equations and the method of least squares estimates for the AR, and usually the method of Durbin estimates for MA. Since AR and MA parameters have been estimated, in the method it is made a new AR filtering of the signal of interest using the estimates of the MA. From this new estimated signal, the new AR and MA estimates of parts from the ARMA process are obtained, and then the power spectral density is estimated. The results depend so much on the spectrum of interest and the parameterization used in the process, but generally the final results were very good. A study describing the main methods of parametric spectral estimation of ARMA processes is also performed in this work. These methods are compared by measuring their accuracy through the relative error and the average coefficient of variation of the parameter estimates / Doutorado / Telecomunicações e Telemática / Doutor em Engenharia Elétrica

[pt] Desenvolveu-se uma técnica de restauração de imagens de defeitos existentes em materiais ferromagnéticos, obtidas pela solução do problema inverso para sinais simulados proveniente do método de Ensaios Não-Destrutivos conhecido como Magnetic Flux Leakage. Baseada em princípios de filtragem ótima, a técnica visa reduzir distorções introduzidas nas imagens pela Inversa MNSL (Minimum-Norm Least-Squares Inverse). Com a aplicação desta técnica, é possível caracterizar mais precisamente diversos tipos de defeitos. / [en] Na image processing technique was developed to restore images of defects in ferromagnetic materials. Such images are obtained by solving the eletromagnetic inverse problem associated with the Magnetic Flux Leakage method of Nondestructive Evaluation. Based upon optimal filtering, the technique aims to reduce distortions introduced by the Minimum-Norm Least-Square Inverse. By applying such technique it is possible to better characterize various defect types.

The paper describes predictive modelling of biomarker data stemming from patients suffering from multiple sclerosis. Improvements of multivariate analyses of the data are investigated with the goal of increasing the capability to assign samples to correct subgroups from the data alone. The effects of different preceding scalings of the data are investigated and combinations of multivariate modelling methods and variable selection methods are evaluated. Attempts at merging the predictive capabilities of the method combinations through voting-procedures are made. A technique for improving the result of PLS-modelling, called bagging, is evaluated. The best methods of multivariate analysis of the ones tried are found to be Partial least squares (PLS) and Support vector machines (SVM). It is concluded that the scaling have little effect on the prediction performance for most methods. The method combinations have interesting properties – the default variable selections of the multivariate methods are not always the best. Bagging improves performance, but at a high cost. No reasons for drastically changing the work flows of the biomarker data analysis are found, but slight improvements are possible. Further research is needed.

Un connecteur électrique automobile est le siège de points de contact électriques soumis à des échauffements supplémentaires par effet Joule provoqués par leurs résistances de contact. La dégradation d’une résistance de contact peut provoquer des échauffements critiques au niveau des points de contact. La présente étude propose deux approches pour détecter les variations anormales des résistances de contact. La première, appliquée à un connecteur de boîtier électronique automobile, repose sur un diagnostic à base de modèle afin de détecter les variations anormales de la résistance de contact par rapport à sa valeur nominale. La seconde approche,appliquée à un connecteur de recharge de véhicule électrique, consiste d’abord à estimer un flux de chaleur lié à l’effet Joule provoqué par la résistance de contact. La valeur de cette dernière est ensuite estimée grâce à la méthode des moindres carrés. Ces deux approches reposent sur des modèles électrothermiques des connecteurs considérés et de leurs environnements. Cette étude présente également le développement de ces modèles, et notamment l’optimisation du maillage d’un fil électrique basée sur la minimisation de la norme H2 de l’erreur entre un modèle maillé d’ordre entier et un modèle analytique de référence d’ordre non entier. / An automotive electrical connector contains electrical contact points subject to additional temperature rises by Joule effect caused by their contact resistances. The deterioration of a contact resistance can cause critical overheating at the contact points. This study proposes two approaches to detect abnormal changes in contact resistance. The first one, applied to an automotive electric case connector, relies on a model-based diagnosis to detect an abnormal variation in the contact resistance with respect to its nominal value. The second one, applied toan electric vehicle charging connector, consists in estimating a heat flow related to the Joule effect caused by the contact resistance. The value of the latter is then estimated using the least squares.These approaches rely on electrothermal models considered connectors and their environments.This study also presents the development of these models, including the optimization of the mesh of an electrical wire based on the minimization of the H2-norm of the error between ameshed integer order model and a reference analytical fractional order model.

Trots att det inom forskningsvärlden propageras för formativ bedömning, kan man i dagsläget notera en mycket stor variation gällande införlivandet av, samt effekter av, formativ bedömning i skolor. Metoder för att kartlägga formativ bedömningspraxis fordras för att kunna särskilja på ”god” respektive ”mindre god” formativ bedömningspraxis. Syftet med föreliggande uppsats var att, med hjälp av en elevenkät och multivariata projektionsmetoder såsom PCA och PLS-DA, kartlägga, och särskilja, formativ bedömningspraxis hos sex olika gymnasieklasser som genomfört kursen kemi 2. Ett sekundärt syfte var även att, med samma verktyg, försöka karakterisera och särskilja frekvenser av olika genomförda undervisningsmoment inom samma kurs och klasser. Studien visade, på ett grafiskt och illustrativt sätt, en stor variation av upplevelser av formativ bedömning inom de tillfrågade klasserna. Vidare visade sig PCA vara ett utmärkt verktyg för att identifiera elevsvar som låg utanför den ”normala” variationen. Genom en PLS-DA-analys påvisades en skillnad i frekvenser av genomförda undervisningsmoment mellan två kommunala och en privat skola – även om dessa resultat bör tolkas med en viss försiktighet.

Due to the invasive nature of quantification techniques, baseline pigment data for coral-dwelling zooxanthellae are not known. In an attempt to develop a model for non-invasive estimation of zooxanthellae pigment concentrations from corals, field samples were taken from Porites rus and P. lutea in Apra Harbor, Guam. In-situ reflectance spectra (R400-R800) from 22 coral colonies were collected. “Coral truthing” was accomplished by extracting corresponding tissue core samples. Subsequent analysis to quantify the concentrations of 6 zooxanthellae pigments (µg cm-2) was performed using HPLC. Trials of multiple linear regressions were attempted (EJ Hochberg) and found inappropriate, despite previous success. The multivariate calibration technique partial least squares regression (PLS-R) is an excellent tool in the case of co-linear variables. Thus, PLS-R was attempted for chlorophyll c2 and peridinin after demonstration of co-linearity. This may be an appropriate approach for development of bio-optical models to estimate zooxanthellae pigment concentrations. Further, the dinoflagellate diagnostic pigment peridinin may be of great value for reef-scale remote sensing of changes in coral status in the future.

This doctoral thesis endeavors to extend probability and statistical models using stochastic differential equations. The described models capture essential features from data that are not explained by classical diffusion models driven by Brownian motion. New results obtained by the author are presented in five articles. These are divided into two parts. The first part involves three articles on statistical inference and simulation of a family of processes related to fractional Brownian motion and Ornstein-Uhlenbeck process, the so-called fractional Ornstein-Uhlenbeck process of the second kind (fOU2). In two of the articles, we show how to simulate fOU2 by means of circulant embedding method and memoryless transformations. In the other one, we construct a least squares consistent estimator of the drift parameter and prove the central limit theorem using techniques from Stochastic Calculus for Gaussian processes and Malliavin Calculus. The second phase of my research consists of two articles about jump market models and arbitrage portfolio strategies for an insider trader. One of the articles describes two arbitrage free markets according to their risk neutral valuation formula and an arbitrage strategy by switching the markets. The key aspect is the difference in volatility between the markets. Statistical evidence of this situation is shown from a sequential data set. In the other one, we analyze the arbitrage strategies of an strong insider in a pure jump Markov chain financial market by means of a likelihood process. This is constructed in an enlarged filtration using Itô calculus and general theory of stochastic processes. / Föreliggande doktorsavhandling strävar efter att utöka sannolikhetsbaserade och statistiska modeller med stokastiska differentialekvationer. De beskrivna modellerna fångar väsentliga egenskaper i data som inte förklaras av klassiska diffusionsmodeller för brownsk rörelse.  Nya resultat, som författaren har härlett, presenteras i fem uppsatser. De är ordnade i två delar. Del 1 innehåller tre uppsatser om statistisk inferens och simulering av en familj av stokastiska processer som är relaterade till fraktionell brownsk rörelse och Ornstein-Uhlenbeckprocessen, så kallade andra ordningens fraktionella Ornstein-Uhlenbeckprocesser (fOU2). I två av uppsatserna visar vi hur vi kan simulera fOU2-processer med hjälp av cyklisk inbäddning och minneslös transformering. I den tredje uppsatsen konstruerar vi en minsta-kvadratestimator som ger konsistent skattning av driftparametern och bevisar centrala gränsvärdessatsen med tekniker från statistisk analys för gaussiska processer och malliavinsk analys.  Del 2 av min forskning består av två uppsatser om marknadsmodeller med plötsliga hopp och portföljstrategier med arbitrage för en insiderhandlare. En av uppsatserna beskriver två arbitragefria marknader med riskneutrala värderingsformeln och en arbitragestrategi som består i växla mellan marknaderna. Den väsentliga komponenten är skillnaden mellan marknadernas volatilitet. Statistisk evidens i den här situationen visas utifrån ett sekventiellt datamaterial. I den andra uppsatsen analyserar vi arbitragestrategier hos en insiderhandlare i en finansiell marknad som förändrar sig enligt en Markovkedja där alla förändringar i tillstånd består av plötsliga hopp. Det gör vi med en likelihoodprocess. Vi konstruerar detta med utökad filtrering med hjälp av Itôanalys och allmän teori för stokastiska processer. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 4: Manuscript. Paper 5: Manuscript.</p>

Les travaux de cette thèse portent sur la stabilité et la stabilisation des systèmes non-linaires représentés par des modèles Takagi–Sugeno (T-S). L'objectif de ces travaux est d'explorer des techniques alternatives aux LMI pour l'analyse et la synthèse de lois de commande dans le cadre non quadratique afin de réduire le conservatisme. Tout d'abord, la stabilisation robuste de systèmes T-S à commutations incertains et perturbés a été considérée. Ainsi, des conditions de stabilisation ont été obtenues sous forme LMI sur la base d'une fonction candidate de Lyapunov à commutations. Puis, une nouvelle approche, pour l'analyse de la stabilité des systèmes non linéaires décrits par des modèles T-S polynomiaux a été proposée. L'objectif est ici d'explorer des techniques alternatives aux LMI dans le cadre non-quadratique. Ainsi, sur la base de travaux préliminaires dévolus à l'analyse de la stabilité via les techniques d'optimisation polynomiale « Sum-Of-Squares » (SOS), l'emploi d'une fonction candidate de Lyapunov polynomiale multiple a été proposée. Celle-ci permet de réduire le conservatisme des approches polynomiales existantes dans la littérature. Enfin, les modèles T-S classiques pouvant-être vus comme un cas particulier des modèles polynomiaux, une méthodologie de synthèse de lois de commande dans le cadre non quadratique est proposée. Celle-ci permet de s'affranchir de paramètres difficiles à obtenir en pratique via les approches LMI ainsi que de fournir une solution globale lorsque celle-ci existe. Néanmoins, à ce jour, des hypothèses fortes de modélisation restent toutefois nécessaires et constituent l'inconvénient majeur des approches SOS. Inconvénient qu'il conviendra de traiter dans des travaux futurs et qui suggèrent donc quelques perspectives à ces travaux. / This thesis deals with the stability and stabilization of nonlinear systems represented by Takagi-Sugeno (T-S) models. The objective of this work is to explore and find out some alternatives to classical LMI conditions in order to reduce the conservatism. First, we focus on robust stabilisation of uncertain switched T-S models. Based on a switched Lyapunov function, the stabilisation conditions are obtained in terms of LMI. Then, a new approach for the stability analysis of polynomial T-S models is proposed. The goal is here to explore alternatives to LMI in the non-quadratic framework. Therefore, an extension of some preliminary result on the stability analysis of polynomial T-S models is proposed by the use of a multiple polynomial Lyapunov function which lead to less conservatism. The stability conditions are given in terms of Sum-of-Squares (SOS) polynomial optimization problem. Finally, classical T-S models being a particular case of polynomial ones, an attempt is done as an alternative to LMI in the non quadratic framework for the design of non-PDC controllers via SOS techniques. Within this framework, global stability may be guaranteed if there exists a solution to the SOS constraints. Moreover, it didn't require unknown parameters in advance like in LMI based non quadratic approaches. However, these SOS based controller design conditions are obtained through a restrictive modelling assumption, suggesting future prospects to this work.

This thesis covers the development of a series of new methods and the application of adaptive filter theory which are combined to produce a generalised adaptive filter system which may be used to perform such tasks as pattern recognition. Firstly, the relevant background adaptive filter theory is discussed in Chapter 1 and methods and results which are important to the rest of the thesis are derived or referenced. Chapter 2 of this thesis covers the development of a new adaptive algorithm which is designed to give faster convergence than the LMS algorithm but unlike the Recursive Least Squares family of algorithms it does not require storage of a matrix with n2 elements, where n is the number of filter taps. In Chapter 3 a new extension of the LMS adaptive notch filter is derived and applied which gives an adaptive notch filter the ability to lock and track signals of varying pitch without sacrificing notch depth. This application of the LMS filter is of interest as it demonstrates a time varying filter solution to a stationary problem. The LMS filter is next extended to the multidimensional case which allows the application of LMS filters to image processing. The multidimensional filter is then applied to the problem of image registration and this new application of the LMS filter is shown to have significant advantages over current image registration methods. A consideration of the multidimensional LMS filter as a template matcher and pattern recogniser is given. In Chapter 5 a brief review of statistical pattern recognition is given, and in Chapter 6 a review of relevant connectionist models. In Chapter 7 the generalised adaptive filter is derived. This is an adaptive filter with the ability to model non-linear input-output relationships. The Volterra functional analysis of non-linear systems is given and this is combined with adaptive filter methods to give a generalised non-linear adaptive digital filter. This filter is then considered as a linear adaptive filter operating in a non-linearly extended vector space. This new filter is shown to have desirable properties as a pattern recognition system. The performance and properties of the new filter is compared with current connectionist models and results demonstrated in Chapter 8. In Chapter 9 further mathematical analysis of the networks leads to suggested methods to greatly reduce network complexity for a given problem by choosing suitable pattern classification indices and allowing it to define its own internal structure. In Chapter 10 robustness of the network to imperfections in its implementation is considered. Chapter 11 finishes the thesis with some conclusions and suggestions for future work.