Dissertations / Theses on the topic 'Modèles de Markov à sauts'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Modèles de Markov à sauts.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Suparman, Suparman. "Problèmes de choix de modèles par simulation de type Monte Carlo par chaînes de Markov à sauts réversibles." Toulouse 3, 2003. http://www.theses.fr/2003TOU30005.
Full textCrudu, Alina. "Approximations hybrides de processus de Markov à sauts multi-échelles : applications aux modèles de réseaux de gènes en biologie moléculaire." Phd thesis, Université Rennes 1, 2009. http://tel.archives-ouvertes.fr/tel-00454886.
Full textCloez, Bertrand. "Comportement asymptotique de processus avec sauts et applications pour des modèles avec branchement." Phd thesis, Université Paris-Est, 2013. http://tel.archives-ouvertes.fr/tel-00862913.
Full textRao, Zusheng. "Etude asymptotique d'un modèle de propagation aléatoire de fissure et filtrage d'une diffusion réfléchie à sauts, observée à travers un processus ponctuel marqué." Aix-Marseille 1, 1993. http://www.theses.fr/1993AIX11030.
Full textZheng, Fei. "Learning and smoothing in switching Markov models with copulas." Thesis, Lyon, 2017. http://www.theses.fr/2017LYSEC066/document.
Full textSwitching Markov Models, also called Jump Markov Systems (JMS), are widely used in many fields such as target tracking, seismic signal processing and finance, since they can approach non-Gaussian non-linear systems. A considerable amount of related work studies linear JMS in which data restoration is achieved by Markov Chain Monte-Carlo (MCMC) methods. In this dissertation, we try to find alternative restoration solution for JMS to MCMC methods. The main contribution of our work includes two parts. Firstly, an algorithm of unsupervised restoration for a recent linear JMS known as Conditionally Gaussian Pairwise Markov Switching Model (CGPMSM) is proposed. This algorithm combines a parameter estimation method named Double EM, which is based on the Expectation-Maximization (EM) principle applied twice sequentially, and an efficient approach for smoothing with estimated parameters. Secondly, we extend a specific sub-model of CGPMSM known as Conditionally Gaussian Observed Markov Switching Model (CGOMSM) to a more general one, named Generalized Conditionally Observed Markov Switching Model (GCOMSM) by introducing copulas. Comparing to CGOMSM, the proposed GCOMSM adopts inherently more flexible distributions and non-linear structures, while optimal restoration is feasible. In addition, an identification method called GICE-LS based on the Generalized Iterative Conditional Estimation (GICE) and the Least-Square (LS) principles is proposed for GCOMSM to approximate any non-Gaussian non-linear systems from their sample data set. All proposed methods are tested by simulation. Moreover, the performance of GCOMSM is discussed by application on other generable non-Gaussian non-linear Markov models, for example, on stochastic volatility models which are of great importance in finance
Bect, Julien. "Processus de Markov diffusifs par morceaux : outils analytiques et numériques." Phd thesis, Université Paris Sud - Paris XI, 2007. http://tel.archives-ouvertes.fr/tel-00169791.
Full textNous introduisons dans la première partie du mémoire la notion de processus diffusif par morceaux, qui fournit un cadre théorique général qui unifie les différentes classes de modèles "hybrides" connues dans la littérature. Différents aspects de ces modèles sont alors envisagés, depuis leur construction mathématique (traitée grâce au théorème de renaissance pour les processus de Markov) jusqu'à l'étude de leur générateur étendu, en passant par le phénomène de Zénon.
La deuxième partie du mémoire s'intéresse plus particulièrement à la question de la "propagation de l'incertitude", c'est-à-dire à la manière dont évolue la loi marginale de l'état au cours du temps. L'équation de Fokker-Planck-Kolmogorov (FPK) usuelle est généralisée à diverses classes de processus diffusifs par morceaux, en particulier grâce aux notions d'intensité moyenne de sauts et de courant de probabilité. Ces résultats sont illustrés par deux exemples de modèles multidimensionnels, pour lesquels une résolution numérique de l'équation de FPK généralisée a été effectuée grâce à une discrétisation en volumes finis. La comparaison avec des méthodes de type Monte-Carlo est également discutée à partir de ces deux exemples.
Cauchemez, Simon. "Estimation des paramètres de transmission dans les modèles épidémiques par échantillonnage de Monte Carlo par chaine de Markov." Paris 6, 2005. http://www.theses.fr/2005PA066572.
Full textKouegou, Kamen Boris. "Grandes déviations dans des modèles de biologie et des épidémies." Thesis, Aix-Marseille, 2019. http://www.theses.fr/2019AIXM0619.
Full textWe are interested in large deviations principle for Markov jump processes and it applications in biology and Eepidemiology. We prove using a new approach the lower bound of the large deviations principle for such general processes and we also write the well known upper bound. We apply these result to a malaria transmission model in epidemiology and give estimate to the exit time from the domain of attraction of the endemic equilibrium. We also apply the approach to obtain large deviations estimates for a model of evolutionary biology which describes the effect of continuous environment changes on the fitness of a given population. Finally we treat a deterministic spatially explicit model of cholera epidemics, propose a stochastic modelling and establish a law of large number. We end by giving large deviations estimates for the stochastic process
Wanderley, Matos de Abreu Thiago. "Modeling and performance analysis of IEEE 802.11-based chain networks." Thesis, Lyon 1, 2015. http://www.theses.fr/2015LYO10030/document.
Full textThe IEEE 802.11 protocol, based on the CMSA/CA principles, is widely deployed in current communications, mostly due to its simplicity and low cost implementation. One common usage can be found in multi-hop wireless networks, where communications between nodes may involve relay nodes. A simple topology of these networks including one source and one destination is commonly known as a chain. In this thesis, a hierarchical modeling framework, composed of two levels, is presented in order to analyze the associated performance of such chains. The upper level models the chain topology and the lower level models each of its nodes. It estimates the performance of the chain in terms of the attained throughput and datagram losses, according to different patterns of channel degradation. In terms of precision, the model delivers, in general, accurate results. Furthermore, the time needed for solving it remains very small. The proposed model is then applied to chains with 2, 3 and 4 nodes, in the presence of occasional hidden nodes, finite buffers and non-perfect physical layer. Moreover, the use of the proposed model allows us to highlight some inherent properties to such networks. For instance, it is shown that a chain presents a performance maximum (with regards to the attained throughput) according to the system workload level, and this performance collapses with the increase of the workload. This represents a non-trivial behavior of wireless networks and cannot be easily identified. However, the model captures this non-trivial effect. Finally, some of the impacts in chains performance due to the IEEE 802.11 mechanisms are analyzed and detailed. The strong synchronization among nodes of a chain is depicted and how it represents a challenge for the modeling of such networks. The proposed model overcomes this obstacle and allows an easy evaluation of the chain performance
Champagnat, Nicolas. "Étude mathématique de modèles stochastiques d'évolution issus de la théorie écologique des dynamiques adaptatives." Phd thesis, Université de Nanterre - Paris X, 2004. http://tel.archives-ouvertes.fr/tel-00091929.
Full textLe, Thanh Trung. "Contribution to deterioration modeling and residual life estimation based on condition monitoring data." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAT099/document.
Full textPredictive maintenance plays a crucial role in maintaining continuous production systems since it can help to reduce unnecessary intervention actions and avoid unplanned breakdowns. Indeed, compared to the widely used condition-based maintenance (CBM), the predictive maintenance implements an additional prognostics stage. The maintenance actions are then planned based on the prediction of future deterioration states and residual life of the system. In the framework of the European FP7 project SUPREME (Sustainable PREdictive Maintenance for manufacturing Equipment), this thesis concentrates on the development of stochastic deterioration models and the associated remaining useful life (RUL) estimation methods in order to be adapted in the project application cases. Specifically, the thesis research work is divided in two main parts. The first one gives a comprehensive review of the deterioration models and RUL estimation methods existing in the literature. By analyzing their advantages and disadvantages, an adaption of the state of the art approaches is then implemented for the problem considered in the SUPREME project and for the data acquired from a project's test bench. Some practical implementation aspects, such as the issue of delivering the proper RUL information to the maintenance decision module are also detailed in this part. The second part is dedicated to the development of innovative contributions beyond the state-of-the-are in order to develop enhanced deterioration models and RUL estimation methods to solve original prognostics issues raised in the SUPREME project. Specifically, to overcome the co-existence problem of several deterioration modes, the concept of the "multi-branch" models is introduced. It refers to the deterioration models consisting of different branches in which each one represent a deterioration mode. In the framework of this thesis, two multi-branch model types are presented corresponding to the discrete and continuous cases of the systems' health state. In the discrete case, the so-called Multi-branch Hidden Markov Model (Mb-HMM) and the Multi-branch Hidden semi-Markov model (Mb-HsMM) are constructed based on the Markov and semi-Markov models. Concerning the continuous health state case, the Jump Markov Linear System (JMLS) is implemented. For each model, a two-phase framework is carried out for both the diagnostics and prognostics purposes. Through numerical simulations and a case study, we show that the multi-branch models can help to take into account the co-existence problem of multiple deterioration modes, and hence give better performances in RUL estimation compared to the ones obtained by standard "single branch" models
Nguyen, Laurent. "Calibration de modèles financiers par minimisation d'entropie relative et modèles avec sauts." Phd thesis, Ecole des Ponts ParisTech, 2003. http://tel.archives-ouvertes.fr/tel-00005766.
Full textLa calibration de modèles financiers par minimisation de lentropie relative a été proposée récemment dans le cadre de la méthode de Monte Carlo. On a étudié la convergence et la stabilité de cette méthode et on a étendu les résultats à des critères plus généraux que lentropie relative. La prise en compte des contraintes sur le sous-jacent assurant labsence dopportunité darbitrage a été abordée sous langle dun problème de moments.
Dans la seconde partie, on a considéré un modèle simple du phénomène de krach en introduisant en particulier des sauts dans la volatilité du sous-jacent. On a calculé le risque quadratique et effectué un développement approché du smile utile pour la calibration.
Finalement, dans la troisième partie, on utilise lentropie relative pour calibrer lintensité des sauts dun modèle de diffusion avec sauts et volatilité locale. La stabilité de la méthode a été prouvée grâce à des techniques de contrôle optimal ainsi quau théorème des fonctions implicites.
Mariton, Michel. "Les systèmes linéaires à sauts markoviens." Paris 11, 1986. http://www.theses.fr/1986PA112288.
Full textDufour, François. "Contribution à l'étude des systèmes linéaires à sauts markoviens." Paris 11, 1994. http://www.theses.fr/1994PA112044.
Full textAbbassi, Noufel. "Chaînes de Markov triplets et filtrage optimal dans les systemes à sauts." Phd thesis, Institut National des Télécommunications, 2012. http://tel.archives-ouvertes.fr/tel-00873630.
Full textJoulin, Aldéric Privault Nicolas. "Concentration et fluctuations de processus stochastiques avec sauts." [S. l.] : [s. n.], 2006. http://tel.archives-ouvertes.fr.
Full textAit, Rami Mustapha. "Approche LMI pour l'analyse et la commande des systèmes à sauts markoviens." Paris 9, 1997. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1997PA090026.
Full textYang, Xiaochuan. "Etude dimensionnelle de la régularité de processus de diffusion à sauts." Thesis, Paris Est, 2016. http://www.theses.fr/2016PESC1073/document.
Full textIn this dissertation, we study various dimension properties of the regularity of jump di usion processes, solution of a class of stochastic di erential equations with jumps. In particular, we de- scribe the uctuation of the Hölder regularity of these processes and that of the local dimensions of the associated occupation measure by computing their multifractal spepctra. e Hausdor dimension of the range and the graph of these processes are also calculated.In the last chapter, we use a new notion of “large scale” dimension in order to describe the asymptotics of the sojourn set of a Brownian motion under moving boundaries
Pchelintsev, Evgeny. "Estimation paramétrique améliorée pour des modèles régressifs observés sous un bruit avec sauts." Rouen, 2012. http://www.theses.fr/2012ROUES041.
Full textThis thesis is devoted to parametric estimation for discret and continuous time regression models which are conditionally Gaussian with respect to a non-observable process. We consider the problem of estimating the unknown parameter using data governed by regression models. We develop improved methods for parameter estimation of regression models compared to least squares estimates. For regression models with Levy noise and Ornstein -- Uhlenbeck noise, we obtain explicit formulas for the minimal gain in mean square accuracy when using shrinkage estimates instead of the least squares estimates. For continuous models, are built improved estimates of the parameters on discrete data. For the model with noise and with jumps, we establish the asymptotic minimaxity of the least squares estimates and of the proposed shrinkage estimates in the sense of robust risk. We also carry on a simulation study of the proposed estimation procedures
Blanchet-Scalliet, Christophette. "Processus à sauts et risque de défaut." Phd thesis, Université d'Evry-Val d'Essonne, 2001. http://tel.archives-ouvertes.fr/tel-00192209.
Full textLa seconde est consacrée à une modélisation du risque de défaut. Nous insistons sur la différence entre l'information liée au défaut de celle du marché sans défaut. Nous établissons des théorèmes de représentation prévisibles pour les martingales dans la filtration élargie.
Boudaren, Mohamed El Yazid. "Modèles graphiques évidentiels." Phd thesis, Institut National des Télécommunications, 2014. http://tel.archives-ouvertes.fr/tel-01004504.
Full textJraifi, Abdelilah. "Analyse numérique de modèles de diffusion-sauts à volatilité stochastique : cas de l'évaluation des options." Thesis, Valenciennes, 2014. http://www.theses.fr/2014VALE0002.
Full textIn the modern economic world, the options contracts are used because they allow to hedge against the vagaries and risks refers to fluctuations in the prices of the underlying assets. The determination of the price of these contracts is of great importance for investors.We are interested in problems of options pricing, actually the European and Quanto options on a financial asset. The price of that asset is modeled by a multi-dimentional jump diffusion with stochastic volatility. Otherwise, the first model considers the volatility as a continuous process and the second model considers it as a jump process. Finally in the 3rd model, the underlying asset is without jump and volatility follows a model CEV without jump. This model allow better to take into account some phenomena observed in the markets. We develop numerical methods that determine the values of prices for these options. We first write the model as an integro-differential stochastic equations system "EIDS", of which we study existence and unicity of solutions. Then we relate the resolution of PIDE to the computation of the option value. This link, which is based on the notion of infinitesimal generators, allows us to use different numerical methods. We therefore introduce the variational equation associated with the PIDE, and drawing on the work of Zhang [106], we show that it admits a unique solution in a weights Sobolev space We focus on the numerical approximation of the price of the option, by treating the problem in a bounded domain. We use the finite elements method of type (P1), and the scheme of Euler-Maruyama, for this serve, on the one hand the finite differences method in time, and on the other hand the method of Monte Carlo and the Quasi Monte Carlo method. For this last method we use of Halton sequences to improve the speed of convergence.We present a comparative study of the different numerical results in many different cases in order to investigate the performance and effectiveness of the used methods
Brezovski, Mathieu. "Inférence bayésienne des modèles à sauts dans la volatilité du sous-jacent des options négociables." Université Louis Pasteur (Strasbourg) (1971-2008), 2005. https://publication-theses.unistra.fr/public/theses_doctorat/2005/BREZOVSKI_Mathieu_2005.pdf.
Full textSince the seminal contributions of Black and Scholes (1973) the range of the possible specifications to model the evolution of an underlying asset has considerably increased and continues to grow. These developments are justified by the need for taking into account the phenomena governing the dynamics of the underlying asset in order to provide theoretical option prices always more in adequacy with those observed. This thesis is devoted to the study of a new class of option pricing models with jumps in the volatility process. A significant place is granted to the estimation of this kind of models. The first two chapters of the thesis enable us to review the essential elements that are used as a basis for the original part of our work. We thus expose the principal results and weaknesses of standard option pricing models (Black and Scholes, stochastic volatility models and jump models), then, we describe the various existing estimation methods with a specific emphasis on the Bayesian approach. In the third chapter, we start from a widespread specification in the financial literature to propose a new model taking into account the existence of discontinuities in volatility process. After having discussed its risk-neutralization and having deduced a pricing formula, we illustrate the impact of jumps on the implied volatility surface. In the fourth chapter we describe an original Bayesian method of estimation based on the observation of the French volatility index, which allows to determine the risk-neutrals parameters of this model. Finally, in the fifth chapter, we expose an extended model where both asset returns and volatility follow jump diffusion processes
Bastide, Paul. "Modèles de processus stochastiques avec sauts sur arbres : application à l'évolution adaptative sur des phylogénies." Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLS370/document.
Full textThis project is aiming at taking a step further in the process of systematic statistical modeling that is occurring in the field of comparative ecology. A way to account for correlations between quantitative traits of a set of sampled species due to common evolutionary histories is to see the current state as the result of a stochastic process running on a phylogenetic tree. Due to environmental changes, some ecological niches can shift in time, inducing a shift in the parameters values of the stochastic process modeling trait evolution. Because we only measure the value of the process at a single time point, for extant species, some evolutionary scenarios cannot be reconstructed, or have some identifiability issues, that we carefully study. We construct an incomplete-data model for statistical inference, along with an efficient implementation. We perform an automatic shift detection, and choose the number of shifts thanks to a model selection procedure, specifically crafted to handle the special structure of the problem. Theoretical guaranties are derived in some special cases. A phylogenetic tree cannot take into account hybridization or horizontal gene transfer events, that are widely spread in some groups of species, such as plants or bacterial organisms. A phylogenetic network can be used to deal with these events. We develop a new model of trait evolution on this kind of structure, that takes non-linear effects such as heterosis into account. Heterosis, or hybrid vigor or depression, is a well studied effect, that happens when a hybrid species has a trait value that is outside of the range of its two parents
Joulin, Aldéric. "Concentration et fluctuations de processus stochastiques avec sauts." Phd thesis, Université de La Rochelle, 2006. http://tel.archives-ouvertes.fr/tel-00115724.
Full textDans la première partie de la thèse, nous explorons le
phénomène de concentration des processus de naissance et de mort. Les différentes approches considérées sont d'une part les inégalités fonctionnelles ainsi que la méthode de
Herbst, et d'autre part l'étude des propriétés du semigroupe associé et des techniques de martingales. En particulier, nous
sommes amenés à introduire diverses notions de courbures de ces processus, analogues discrets du critère de courbure de Bakry-Emery dans le cadre des processus de diffusion.
Dans la deuxième partie de la thèse, nous étudions le
comportement du processus supremum d'une intégrale stable stochastique en établissant des inégalités maximales que nous appliquons à des problèmes de temps de passage de
processus symétriques stables. Enfin, nous démontrons un principe de domination convexe pour des intégrales stochastiques brownienne et stable corrélées.
Ben, Mabrouk Mohamed. "Modèles de Markov triplets en restauration des signaux." Phd thesis, Institut National des Télécommunications, 2011. http://tel.archives-ouvertes.fr/tel-00694128.
Full textGagnon, Sébastien. "Modèles de Markov cachés à haute précision dynamique." Mémoire, Université de Sherbrooke, 2016. http://hdl.handle.net/11143/8996.
Full textRobinson, Sean. "Applications en bioinformatique avec des modèles de Markov." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAS017/document.
Full textIn this thesis we present four applications in bioinformatics with Markov models. Such models are especially popular since the Markov structure allows for complex conditional independences to be modelled while still allowing for efficient inference. We achieve a variety of aims, ranging from alignment, classification, segmentation and quantification, through inference in different types of Markov models. In this way we show that Markov models can be used to generate new knowledge in diverse applications relating to multiple domains of biological research
Tordeux, Antoine. "Étude de processus en temps continu modélisant l'écoulement de flux de trafic routier." Phd thesis, Université Paris-Est, 2010. http://tel.archives-ouvertes.fr/tel-00596941.
Full textDjama, Zahir. "Approche multi modèles à sauts markoviens et fusion multi capteurs pour la localisation d'un robot mobile." Paris 12, 2001. http://www.theses.fr/2001PA12A001.
Full textFusion and filtering techniques currently used for the localization of a mobile robot present two main drawbacks. The first one concerns the fact that no a priori reliable information on the input and the measurement noise covariance is generaily available. The second one is tied to the fact that the process of localization is often modelled under the form of a unique model leading to the introduction of modelling errors that degrade the quality of the filtering. The work presented in this thesis presents two contributions. The first one consists in taking into account the existence of several regimes in the localization process. This one is modelled under the form of a Markovian hybrid process both from state and observation procesess point of view. The second contribution consists in proposing an on-une adaptative estimation of statistical parameters such as state and observation noise variances along with an optimal management of observations. The fusion of data is performed by Kalman filters of adaptive linear type for linear process and of adaptive extended type for non linear process. This approach has been validated in simulation on a robot equipped with an odometer, two telemeters perpendicularly displayed and a compas. In order to show its efficiency, a comparative analysis of its performance with respect to existing approaches is presented. Thus, gains values on accuracy obtained hy this approach compared to classical filters are 2 on translation and 2 on orientation
Laporte, Francis. "Méthode d'inférence utilisant la vraisemblance empirique basée sur l'entropie pour les modèles de diffusion avec sauts." Master's thesis, Université Laval, 2019. http://hdl.handle.net/20.500.11794/33909.
Full textWith the advent of increasingly sophisticated models for modeling stock market returns, the classical maximum likelihood method for inferring parameters is generally no longer applicable since, for example, the density function has no closed form or very difficult to calculate numerically. In the literature, inference by the method of moments (MM) is therefore generally suggested. In this master’s thesis, a more efficient inference method, the maximum empirical entropy likelihood (MEEL), is proposed for two particular cases of the Lévy process, namely the Merton and Tsay models. First, a review of some models developed in the past is done. The flaws of the geometric Brownian motion are presented to justify the use of more sophisticated models. Then, the two models, Merton and Tsay, and their properties are presented in more detail. Subsequently, there is a comparative analysis between the effectiveness of the MEEL and the MM; an example with real data is also presented. Finally, two approaches to pricing derivatives are presented.
Mikou, Mohammed. "Options américaines dans les modèles exponentiels de Lévy." Phd thesis, Université Paris-Est, 2009. http://tel.archives-ouvertes.fr/tel-00628448.
Full textAupetit, Sébastien. "Contributions aux Modèles de Markov Cachés : métaheuristiques d'apprentissage, nouveaux modèles et visualisation de dissimilarité." Phd thesis, Université François Rabelais - Tours, 2005. http://tel.archives-ouvertes.fr/tel-00168392.
Full textde métaheuristiques biomimétiques classiques (les algorithmes génétiques, l'algorithme de fourmis artificielles API et l'optimisation par essaim particulaire) au problème de l'apprentissage de MMC. Dans la
deuxième partie, nous proposons un nouveau type de modèle de Markov caché, appelé modèle Markov caché à substitutions de symboles (MMCSS). Un MMCSS permet d'incorporer des connaissances a priori dans le processus d'apprentissage et de reconnaissance. Les premières expérimentations de ces modèles sur des images démontrent leur intérêt. Dans la troisième partie, nous proposons une nouvelle méthode de représentation de dissimilarité appelée matrice de scatterplots pseudo-euclidienne (MSPE), permettant de mieux comprendre les interactions entre des MMC. Cette MSPE est construite à partir
d'une technique que nous nommons analyse en composantes principales à noyau indéfini (ACPNI). Nous terminons par la présentation de la bibliothèque HMMTK, développée au cours de ce travail. Cette dernière intègre des mécanismes de parallélisation et les algorithmes développés au cours de la thèse.
Lehéricy, Luc. "Estimation adaptative pour les modèles de Markov cachés non paramétriques." Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLS550/document.
Full textDuring my PhD, I have been interested in theoretical properties of nonparametric hidden Markov models. Nonparametric models avoid the loss of performance coming from an inappropriate choice of parametrization, hence a recent interest in applications. In a first part, I have been interested in estimating the number of hidden states. I introduce two consistent estimators: the first one is based on a penalized least squares criterion, and the second one on a spectral method. Once the order is known, it is possible to estimate the other parameters. In a second part, I consider two adaptive estimators of the emission distributions. Adaptivity means that their rate of convergence adapts to the regularity of the target distribution. Contrary to existing methods, these estimators adapt to the regularity of each distribution instead of only the worst regularity. The third part is focussed on the misspecified setting, that is when the observations may not come from a hidden Markov model. I control of the prediction error of the maximum likelihood estimator when the true distribution satisfies general forgetting and mixing assumptions. Finally, I introduce a nonhomogeneous variant of hidden Markov models : hidden Markov models with trends, and show that the maximum likelihood estimators of such models is consistent
Bloch, Daniel. "Modèles de diffusion à sauts affine et quadratique : application aux nouvelles options exotiques dans les marchés actions et hybrides." Paris 6, 2006. http://www.theses.fr/2006PA066635.
Full textThis thesis is concerned with the pricing of exotic options within an affine quadratic jump diffusion model. In this case the computational difficulties can be reduced to solving a system of Riccati equations a number of times and performing a numerical integration using the resulting values via the FFT technique. We then present the variance swap contract and explain the reasons why it became a traded underlying. Since the variance swap contract is just a forward on the annualised realised variance we choose to infer its dynamic from the dynamic of the stock price. We therefore make the variance swap the new underlying and diffuse it over time in order to price options on the quadratic variation and more generally derivatives on the volatility. The properties of the affine-quadratic model allow us in some special cases to recover closed-form solutions. To conclude we extend the approach to the hybrid markets and consider the equity-rate and equity-credit products
Zakoian, Jean-Michel. "Modèles autorégressifs à seuil de séries chronologiques." Paris 9, 1990. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1990PA090014.
Full textAlvarez, Alexander. "Modélisation de séries financières, estimation, ajustement de modèles et test d'hypothèses." Toulouse 3, 2007. http://www.theses.fr/2007TOU30018.
Full textBinsztok, Henri. "Apprentissage de modèles Markoviens pour l'analyse de séquences." Paris 6, 2007. http://www.theses.fr/2007PA066568.
Full textInitially, Machine Learning allowed to learn models from labeled data. But, for numerous tasks, notably for the task of user modeling, if the available quantity of data is potentially without limit, the quantity of labeled data is almost nonexistent. Within the framework of this thesis, we are interested in the unsupervised learning of sequence models. The information of sequence constitutes the first level of structured data, where the data are no more simple vectors of characteristics. We propose approaches that we apply to the automatic learning of Hidden Markov Models ( HMMs) and Hierarchical HMMs (HHMMs). Our purpose is to learn simultaneously the structure and the parameters of these Markovian Models, to minimize the quantity of prior information necessary to learn them
Ponomarev, Denis. "Modèles markoviens pour le mélange des poudres en mélangeur statique." Toulouse, INPT, 2006. http://ethesis.inp-toulouse.fr/archive/00000359/.
Full textMixing of powders is important in many different industries as it adds significant value to the product. The research on this operation in industry goes through simulation of continuous mixing and mixing in general. The objective of this thesis is a development of a unified model representing samples of the particulate flow in different types of static mixers operating with materials that tend to segregate. One and two dimensional models of static mixers based on the theory of Markov chains are developed and the procedure of model parameter identification is proposed. The results of simulation have shown how the value of transition probabilities and initial distribution of the components influence mixture quality. The experimental results have proved the adequacy of the developed models. It has been shown that there is an optimum number of material passages through the mixer that gives maximal mixture quality while mixing components that tend to segregate
Mercier, Sophie. "Modèles et optimisation de politiques de maintenance de systèmes." Université de Marne-la-Vallée, 2000. http://www.theses.fr/2000MARN0087.
Full textLeocard, Stéphanie. "Modèles probabilistes du balayage sélectif et auto-stop génétique." Aix-Marseille 1, 2009. http://www.theses.fr/2009AIX11024.
Full textIn this manuscript, several topics of probability applied to genetics have been studied. First, we are interested in the hitchhiking effect in case of selective sweep. We consider a high but finite selective advantage and we describe an approximate distribution for the genealogy of an arbitrary number of partially linked neutral genes in a sample of n individuals. Then we obtain an approximate distribution for the number of hitchhiking alleles at the end of the selective sweep for this sample. In a second time, we make the selective advantage go to infinity, so that the selective sweep is instantaneous. We study the evolution of the coalescent tree and the ancestral recombination graph as the distance from the selected locus increases. Then we establish some properties of the “look-down” construction in case of selection. Finally, in the last chapter, independent from the first two ones, we construct a detection method for horizontal gene transfers. We implemented a program that, from a reference species tree and a gene alignment, gives the most probable transfers and the corresponding gene tree with its log-likelihood
Alain, Pierre. "Contributions à l'évaluation des modèles de langage." Rennes 1, 2007. http://www.theses.fr/2007REN1S003.
Full textThis work deals with the evaluation of language models independently of any applicative task. A comparative study between several language models is generally related to the role that a model has into a complete system. Our objective consists in being independant of the applicative system, and thus to provide a true comparison of language models. Perplexity is a widely used criterion as to comparing language models without any task assumptions. However, the main drawback is that perplexity supposes probability distributions and hence cannot compare heterogeneous models. As an evaluation framework, we went back to the definition of the Shannon's game which is based on model prediction performance using rank based statistics. Our methodology is able to predict joint word sequences that are independent of the task or model assumptions. Experiments are carried out on French and English modeling with large vocabularies, and compare different kinds of language models
Yahiaoui, Meriem. "Modèles statistiques avancés pour la segmentation non supervisée des images dégradées de l'iris." Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLL006/document.
Full textIris is considered as one of the most robust and efficient modalities in biometrics because of its low error rates. These performances were observed in controlled situations, which impose constraints during the acquisition in order to have good quality images. The renouncement of these constraints, at least partially, implies degradations in the quality of the acquired images and it is therefore a degradation of these systems’ performances. One of the main proposed solutions in the literature to take into account these limits is to propose a robust approach for iris segmentation. The main objective of this thesis is to propose original methods for the segmentation of degraded images of the iris. Markov chains have been well solicited to solve image segmentation problems. In this context, a feasibility study of unsupervised segmentation into regions of degraded iris images by Markov chains was performed. Different image transformations and different segmentation methods for parameters initialization have been studied and compared. Optimal modeling has been inserted in iris recognition system (with grayscale images) to produce a comparison with the existing methods. Finally, an extension of the modeling based on the hidden Markov chains has been developed in order to realize an unsupervised segmentation of the iris images acquired in visible light
Ledoux, James. "Modèles markoviens : sur la caractérisation de l'agrégation faible et sur les modèles structurels pour l'évaluation de la sûreté de fonctionnement du logiciel." Rennes 1, 1993. http://www.theses.fr/1993REN10180.
Full textForbes, Florence. "Modèles markoviens de ressources partagées." Phd thesis, Université Joseph Fourier (Grenoble), 1996. http://tel.archives-ouvertes.fr/tel-00004991.
Full textSimenhaus, François. "Marches aléatoires en milieux aléatoires : étude de quelques modèles multidimensionnels." Paris 7, 2008. http://www.theses.fr/2008PA077148.
Full textThis dissertation is devoted to different models of random walks in random environments; it is made of 5 Chapters. Chapter 1 and 4 are surveys of literature devoted, respectively, to i. I. D model and models where environments is given by a percolation. In Chapter 2 we give a characterization of the class of walks admitting an asymptotic direction in the case of i. I. D. Model. In Chapter 3 we study a model of continuous time random walk in a random i. I. D. Environment. Chapter 5 is devoted to a model of walk delayed by the clusters of a site subcritical percolation
Heutte, Natacha. "Modèles semi-markoviens et données de survie." Paris 5, 2001. http://www.theses.fr/2001PA05S013.
Full textModels for survival data including a categorized quality of life index is proposed. The model is intended to take into account the effect of endogenous and exogenous factors both on the duration of survival and the quality of life. Endogenous factors are for example biological measurements or genetical specifications, while exogenous ones are environmental factors. The proposed models are semi-parametric and based on semi-markov processes. Time may be continuous or discrete depending on the type of the data. The general framework of all preexisting models is sketched. Estimators are derived, as well as their asymptotic properties, and algorithms and programs are given to compute them explicitly. They are exemplified on real data on AIDS and cancer patients, and on simulations. Those models are presented in a biomedical context but can be useful in any field where durations together with multistate processes are involved
Karami, Abir Béatrice. "Modèles Décisionnels d'Interaction Homme-Robot." Caen, 2011. http://www.theses.fr/2011CAEN2077.
Full textThis thesis is focused on decision models for human-robot interaction based on Markovian Decision Processes. First, we propose an augmented decision model that allows a companion robot to act considering estimated human intentions. This model addresses the problem of estimating the intention of the human by observing his actions. We proposed to simulate the behavior of a human to build a library of human action values toward his possible intentions. These values are integrated into the augmented Partially Observable Markov Decision Process (POMDP). Second, we present a coactive decision model that allows a robot in collaboration with a human to choose his behavior according to the progress of the shared task. This model is based on an augmented POMDP and allows the robot to act coactively to encourage the human actions and to perform the task in harmony with him. Third, we also propose a unified model for different types of human-robot interactions where the robot analyzes the needs of the human and acts accordingly. To overcome the complexity of POMDPs, the unified model divides the problem into several parts, the first estimates the human intention with a hidden Markov model (HMM) and another is responsible for choosing the corresponding type of interaction (collaboration, assistance, cooperation) using a Markov Decision Process (MDP). Finally, we propose a model that alternates between verbal interaction to infer the preference of the human using queries and non-verbal interaction in which preferences are estimated by observing the human actions. This model switches back to the verbal interaction when an ambiguity about the preferences is detected
Garivier, Aurélien. "Modèles contextuels et alphabets infinis en théorie de l'information." Paris 11, 2006. http://www.theses.fr/2006PA112192.
Full textThis thesis explores some contemporary aspects of information theory, from source coding to issues of model selection. We first consider the problem of coding memoryless sources on a countable, infinite alphabet. As it is impossible to provide a solution which is both efficient and general, two approaches are considered: we first establish conditions under which the entropic rate can be reached, and we consider restricted classes for which tail probabilities are controlled. The second approach does not set any condition on the sources but provides a partial solution by coding only a part of the information - the pattern - which captures the repetitions in the message. In order to study more complex processes, we come back to the case of finite memory sources on a finite alphabet : it has given rise to many works and efficient algorithms like the Context Tree Weighting (CTW) Method. We show here that this method is also efficient on anon-parametric class of infinite memory sources: the renewal processes. We show then that the ideas on which CTW is based lead to a consistent estimator of the memory structure of a process, when this structure is finite. In fact, we complete the study of the BIC context tree estimator for Variable Length Markov Chains. In the last part, it is shown how similar ideas can be generalized for more complex sources on a (countable or not) infinite alphabet. We obtain consistent estimators for the order of hidden Markov models with Poisson and Gaussian emission
Mares, Ondrej. "Modèles stochastiques de transition et de structures vitreuses." Paris 6, 2003. http://www.theses.fr/2003PA066207.
Full text