Dissertations / Theses on the topic 'Statistiques appliquées'
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 'Statistiques appliquées.'
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.
Benomar, Othman. "Sismologie stellaire : méthodes statistiques appliquées aux étoiles de type solaire." Paris 11, 2010. http://www.theses.fr/2010PA112309.
Full textThis thesis focused on the analysis of the power spectra of solar-like stars observed by the Corot and Kepler missions. The analysis of these stars proved to be difficult, their pulsations being of low intensity and because of the intrinsic properties of the pulsations of some of them. Robust methods of inference of the properties of the stellar modes have been developed. They are based on a Bayesian approach, using either Markov Chain Monte Carlo sampling methods (MCMC) or maximum a posteriori (MAP) methods. Such approaches are based on a mathematical formalism taking into account prior information of the studied object. Three issues are addressed: How to identify the stellar pulsation modes with time series at low signal noise ratio? How to extract the parameters of the stellar pulsations of modes on a large set of stars? What is the internaI structure of the observed stars? Capabilities of my methods are demonstrated using simulations and observations of three stars: the Sun, HD49933 and HD181420. The adjustment of the power spectrum of HD49933 and HD181420 by MCMC method allows us to identify the modes with a high degree of confidence compares to more traditional methods of signal analysis. Applying the developed algorithm on these data also shows the reliability of an automatic identification method based on a MAP approach. Finally, models of internal structure of HD49933 and HD181420 using the seismic observations are presented: for instance we show that these stars have a convective core
Gaeta, Michel. "Les statistiques d'ordre supérieur appliquées à la séparation de sources." Grenoble INPG, 1991. http://www.theses.fr/1991INPG0064.
Full textMercier, Ludovic. "Séries temporelles chaotiques appliquées à la finance problèmes statistiques et algorithmiques." Paris 9, 1998. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1998PA090049.
Full textRaillard, Nicolas. "Modélisation du comportement extrême de processus spatio-temporels. Applications en océanographie et météorologie." Phd thesis, Université Rennes 1, 2011. http://tel.archives-ouvertes.fr/tel-00656468.
Full textPirrachio, Romain. "Méthodologies d'inférence causale appliquées à l'analyse de données observationnelles de réanimation." Paris 7, 2012. http://www.theses.fr/2012PA077023.
Full textBenefit of Intensive Care Unit cannot be evaluated by performing randomized controlled trials, and therefore rely on observational studies. Standard analyses cannot be used to draw causal conclusions. Hence, specific statistical methods have been developed over the past decades in order to draw causal conclusions from observational data. Most of these methods rely on modelling the probability of treatment, the most popular being the propensity score. The first part of the thesis is dedicated to a review of the literature focussing on the use of thèse methods in the major Anesthesiology and Critical Care journals. We then focussed on the potential limitations related to the use of the propensity score. We first evaluated the performance of this method in the context of small sample size or low prevalence of the treatment. We then evaluated the generalized propensity approach as an option for the estimation of treatments with more than two categories. The presence of time-dependant confoundings cannot be handled by propensity score methods. Therefore, we evaluated thé use of marginal structural model as an alternative to propensity score in case of time-dependant confoundings. Finally, because all methods based on probability-of-treatment models rely on the hypothesis of absence of unmeasured confoundings, the last section of this thesis was dedicated to the use of instrumental variable approach when the presence of unmeasured confoundings can be suspected
Richard, Frédéric. "Modèles et analyses statistiques de l'image biomédicale." Habilitation à diriger des recherches, Université René Descartes - Paris V, 2009. http://tel.archives-ouvertes.fr/tel-00522704.
Full textAugustin, Lefèvre. "Méthodes d'apprentissage appliquées à la séparation de sources mono-canal." Phd thesis, École normale supérieure de Cachan - ENS Cachan, 2012. http://tel.archives-ouvertes.fr/tel-00764546.
Full textCouteron, Pierre. "Statistiques spatiales appliquées à l'étude de la végétation : un lien entre structure et processus dynamiques." Habilitation à diriger des recherches, Université Montpellier II - Sciences et Techniques du Languedoc, 2005. http://tel.archives-ouvertes.fr/tel-00090679.
Full textRotty, Marie-Caroline. "Modélisations statistiques de données de cohortes de patients suivis à domicile." Thesis, Montpellier, 2019. http://www.theses.fr/2019MONTS055.
Full textSleep Apnea syndrome (SAS) is a common sleep disorder with a prevalence ranging from 5.9% to 79.2% in European general populations over 35 years of age old but remains under diagnosed. Continuous positive airway Pressure (CPAP) remains the treatment reference for moderate to severe forms of SAS in 2019. CPAP benefits depend on sufficient patient adherence, which is conditioned by multiple factors, including interface related side effects (leakage and their consequences, among others).In this thesis, we demonstrate the feasibility of OSA screening with Jawac® signaling in the general population. The analysis of the long-term CPAP-treated patient results in the interface-vent cohort highlighted that side-effects reported by these patients (some associated with leakage) are associated with compliance and sleepiness, while machine data (residual apnea/hypopnea index and machine leakage) are not.This lack of association between machine data (that could be monitored as proposed in the ATS 2013 algorithm) and patient compliance was confirmed for the patient subgroup of the cohort respecting the machine criteria and interface equipment of the ATS 2013 algorithm.Collaborative work has also made it possible to address the issue of leaks reported by the machine under two other aspects. First of all, it has been shown that the auto- or constant mode of the CPAP has no effect on the level of the leak reported by the machine (while the oro-nasal interface and the pressure level machine have an impact). By studying determinants of leaks during a night of polygraphic recording, the heterogeneity of unintentional leak determinants at the individual was demonstrated. Finding a consensus for how to quantify leak patterns remains a challenge in the field
Soudain-Pineau, Mickaël. "Statistiques appliquées à la physiologie du sport dans l’exploration des variables influençant la performance chez les cyclistes." Reims, 2008. http://theses.univ-reims.fr/exl-doc/GED00000981.pdf.
Full textFirst, we studied a population of 112 cyclists divided into three levels amateurs. These subjects carried out an incremental test by three minute old stage. Anthropometric variables like physiological and physical were studied at two times, at the lactic threshold and at the maximum of the for each individual. We used discriminant analysis to obtain with the lactic threshold a linear discriminant function and in max a quadratic discriminant function made up of the most significant variables. Then, for 213 professional cyclists, we had the values of several hormones before and after an incremental test. We also had physical and physiological parameters for each athlete. We studied the behavior of these hormones and the impact of the physical and physiological parameters on the excepted values. The multiple regression analysis allowed to establish a linear model made up of the most significant parameters explaining the power. Finally, the study of the lactatemy taken at an incremental exercise and for 10 minutes after exercice for a population of professional cyclists allowed, using an existing model, to modeling the blood lactate reaction. This function describing the blood lactate evolution throughout all exercice duration allow to carry out a simulation on the return to a basal level
Faucon, Jean-Christophe. "Etudes statistiques et des relations structure-écotoxicité appliquées aux données écotoxicologiques d'un ensemble hétérogène de substances nouvelles." Caen, 1998. http://www.theses.fr/1998CAEN4002.
Full textSedki, Mohammed. "Échantillonnage préférentiel adaptatif et méthodes bayésiennes approchées appliquées à la génétique des populations." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2012. http://tel.archives-ouvertes.fr/tel-00769095.
Full textGaudoin, Olivier. "Outils statistiques pour l'évaluation de la fiabilité des logiciels." Grenoble 1, 1990. http://www.theses.fr/1990GRE10138.
Full textSalipante, Florian. "Mise au point de méthodologies statistiques appliquées à des données issues de la génomique : puces à ADN, ChIP-chip et ChIP-Seq." Thesis, Montpellier 2, 2011. http://www.theses.fr/2011MON20176/document.
Full textResearch in Genomics produces very huge data which still increase with technology. Statistics is becoming essential to treat this amount of information. These new kind of data represent a great challenge in data analysis because of the great dimensions, the important background and the absence of "golden standard" which could allow to validate the obtained results. During this PhD thesis, we focused on statistical analysis for three kinds of data: DNA microarray, ChIP-chip and ChIP-Seq. For each one, a new approach have been proposed. For DNA microarrays, the GAGG method allows to detect differentially expressed genes and to cluster them by profile types. To do so, a Genetic Algorithm is used in order to optimize two criteria related to two nearby methods of PCA and $k$-means. In the case of ChIP-chip data, the POTChIPS method have been proposed. It allows to detect the binding sites of a protein of interest (a transcription factor e.g.) along the genome. In this method a peak extraction i realized then a significant threshold is obtained from a POT modelization. Finally, for ChIP-Seq data, the goal is the same that the one of chIP-chip, i.e., to find on DNA the binding sites of a protein of interest. The POTSeq method is an adaptation of POTChIPS for ChIP-Seq.La méthode POTSeq, mise au point au cours de cette thèse, est une adaptation de POTChIPS aux données de ChIP-Seq
Poirier, Bertrand. "Analyses statistiques appliquées aux contrôles d'activité in vitro des vaccins humains : impact dans le projet d'accréditation d'un laboratoire de contrôles de médicaments biologiques (standard ISO 17025)." Lyon 1, 2002. http://www.theses.fr/2002LYO10015.
Full textCembrzynski, Thierry. "Conception et réalisation d'outils statistiques et d'intelligence artificielle pour l'aide à la planification du réseau de transport d'Electricité de France." Paris 9, 1988. https://bu.dauphine.psl.eu/fileviewer/index.php?doc=1988PA090015.
Full textBousquet, Nicolas. "Analyse bayésienne de la durée de vie de composants industriels." Paris 11, 2006. http://www.theses.fr/2006PA112237.
Full textThis thesis takes place in a context when Bayesian techniques are sued to estimate magnitudes in industrial reliability and durability, when lifetime is represented by a parametric statistical model. We firstly focus our study around the Weibull distribution, modelling components submitted to aging. A competing risk model between aging and accidental failure is developed. Bayesian technics are used because lifetime data are typically in a few number and contain censored values. Besides, some expert knowledge is provided. Thus, a prior family is elicited such that its calibration is simple for an industrial analyst. Especially, the location of the subjective information with respect to the objective data information is a point of interest. The main contribution of the thesis is the definition and the study of the DAC criterion, which measures a possible discrepancy between a prior and available data, in a larger setting than reliability. DAC improves the existing techniques, has some default calibration properties and constitutes a helpful technique in the toolkit of the Bayesian industrial analyst. Finally, a review of importance sampling approaches - which appear to be especially adapted to Bayesian computation in our industrial settings - ends this thesis work, whosefirst aim is to propose methodological avenues for engineers
Browne, Thomas. "Regression models and sensitivity analysis for stochastic simulators : applications to non-destructive examination." Thesis, Sorbonne Paris Cité, 2017. http://www.theses.fr/2017USPCB041/document.
Full textMany industries perform non-destructive examination (NDE) in order to ensure the integrity of some important components, where the goal is to detect a hypothetical flaw before the occurrence any non desirable event (breaks, leaks etc...). As the NDE environment is highly random, it is legitimate to wonder about the reliability of such procedures. To this effect, their efficiencies are quantified through their probabilities of detection. They are modelled by increasing curves, functions of the flaw size to detect: Probability of Detection-curves (PoD-curves). This thesis aims to introduce tools in order to both build and study PoD-curves through parsimonious strategies. Our approach mainly consists in modelling these curves by random cumulative distribution functions. From this point, it becomes necessary to define relevant objects to quantify their probability distributions. Kriging-based estimators for these values are introduced. Techniques are developed in order to carry out sequential strategies for both the improvement of the PoD-curve estimators and optimization problems. Besides, in order to quantify the influence of input parameters over the quality of detection, functional sensitivity analysis indices are provided. Their respective estimators are also introduced
Barré, Anthony. "Analyse statistique de données issues de batteries en usage réel sur des véhicules électriques, pour la compréhension, l’estimation et la gestion des phénomènes de vieillissement." Thesis, Grenoble, 2014. http://www.theses.fr/2014GRENT064/document.
Full textDue to different reason The electrical vehicle market is undergoing important developments. However the limits associated with performance represent major drawbacks to increase the sales even more. The batteries performance and lifetime are the main focus of EV users. Batteries are subject to performance loss due to complex phenomena implying interactions between the different life conditions of the battery. In order to improve the understanding and estimation of battery aging, the studies were based on datasets from real use ev batteries. More precisely, this study consists in the adaptation and application of statistical approaches on the available data in order to highlight the interactions between variables, as well as the creation of methods for the estimation of battery performance. The obtained results allowed to illustrate the interests of a statistical approach. For example the demonstration of informations contained in the signals coming from the battery which are useful for the estimation of its state of health
Minvielle-Larrousse, Pierre. "Méthodes de simulation stochastique pour le traitement de l’information." Thesis, Pau, 2019. http://www.theses.fr/2019PAUU3005.
Full textWhen a quantity of interest is not directly observed, it is usual to observe other quantities that are linked by physical laws. They can provide information about the quantity of interest if it is able to solve the inverse problem, often ill posed, and infer the value. Bayesian inference is a powerful tool for inversion that requires the computation of high dimensional integrals. Sequential Monte Carlo (SMC) methods, a.k.a. interacting particles methods, are a type of Monte Carlo methods that are able to sample from a sequence of probability densities of growing dimension. They are many applications, for instance in filtering, in global optimization or rare event simulation.The work has focused in particular on the extension of SMC methods in a dynamic context where the system, governed by a hidden Markov process, is also determined by static parameters that we seek to estimate. In sequential Bayesian estimation, the determination of fixed parameters causes particular difficulties: such a process is non-ergodic, the system not forgetting its initial conditions. It is shown how it is possible to overcome these difficulties in an application of tracking and identification of geometric shapes by CCD digital camera. Markov Monte Carlo Chain (MCMC) sampling steps are introduced to diversify the samples without altering the posterior distribution. For another material control application, which mixes static and dynamic parameters, we proposed an original offline approach. It consists of a Particle Marginal Metropolis-Hastings (PMMH) algorithm that integrates Rao-Blackwellized SMC, based on a bank of interacting Ensemble Kalman filters.Other information processing works has been conducted: particle filtering for atmospheric reentry vehicle tracking, 3D radar imaging by sparse regularization and image registration by mutual information
Ribereau, Pierre. "Quelques contributions à la statistique théorique et appliquée." Paris 6, 2005. http://www.theses.fr/2005PA066164.
Full textGalharret, Jean-Michel. "Statistique bayésienne appliquée en sciences humaines et sociales." Thesis, Nantes, 2019. http://www.theses.fr/2019NANT4062.
Full textThis work focuses on two issues: the first concerns mediation analysis and the second concerns dating in archaeology in the presence of outliers. Concerning the mediation analysis, I have developed a non-parametric approach to test the existence of direct effect in binary response causal mediation. Second, prior distributions are compared in Bayesian inference for causal mediation with binary response. Third, the B-priors are adapted to mediation models from psychological research and a moderation test based on the Bayes factor is proposed. Concerning dating in archaeology, a methodology is developed to estimate ages in the presence of outliers. This is based on an existing Bayesian model (Event Model) from which an outlier detection test is developed. This methodology is applied in the combination of measurements and in the estimation of an OSL age (Optically Stimulated Luminescence)
Cornec, Matthieu. "Inégalités probabilistes pour l'estimateur de validation croisée dans le cadre de l'apprentissage statistique et Modèles statistiques appliqués à l'économie et à la finance." Phd thesis, Université de Nanterre - Paris X, 2009. http://tel.archives-ouvertes.fr/tel-00530876.
Full textLusinchi, Dominic. "La statistique appliquée : usage et signification dans les sciences sociales : essai de recherche méthodologique basé sur des études de cas aux États-Unis." Paris 8, 2008. http://octaviana.fr/document/137824084#?c=0&m=0&s=0&cv=0.
Full textThis study examines the statistical tools available to social science practitioners. The last 30 years have witnessed a considerable development in statistical techniques, both numerical and graphical. This is in large part a result of the application of statistical methods to an ever increasing number of fields, and also of the emergence, barely 20 years ago, of the personal computer. Using real data from surveys and other studies conducted in the U. S. , this research will show how important problems that arise in empirical data can be tackled by relying on well-known as well as relatively recent statistical techniques. Applied statistics is not simply an array of procedures; it is above all a way of thinking about empirical data, specifically how to discriminate between real and chance effects. This research endeavors to show that the role of applied statistics is to reveal both the meaning of the data and their limitations. The application of statistical methods to empirical data can often act as a catalyst to stimulate the sociological imagination
Gerville-Réache, Léo. "Analyse statistique de modèles probabilistes appliqués aux processus sociaux." Bordeaux 1, 1998. http://www.theses.fr/1998BOR10606.
Full textBergman, Jacques. "La méthode des modules fonctionnels appliquée à la détermination du coût d'un investissement : cas des unités pétrochimiques." Dijon, 1987. http://www.theses.fr/1987DIJOE002.
Full textThe use of advanced statistical multivariate "data-analysis" technique for a large set of process data, obtained by a computer design procedure and by private IFP data, has led to a new rapide method of capital cost estimated for chemical plants. In the past a whole family of shortcut methods, using a modular approach has been conceived. Nevertheless the undertaking of an integrated project with extensive use of techniques such as principal component analysis, principales factor analysis and multiple linear regression together with modular principles and cost structure philosophy, seems not to have been made thusfar. We chose, as in the detailed IFP evaluating method, to maintain the modular approach model, changing only the conception of the chemical processing block. There a functional approach led to the concept of functional modules based mainly on the basic concept of unit operation. Otherwise for each functional module prior statistical analysis highligted the appropriate set of correlating variables regarding more than one hundred possible configurations. This procedure ensured a remarkable flexibility when the final correlating model was developped. The general nature of this work is heightened by a set of factors, ensuring a choice of materials compatible with the involved products specification. The application of these elements to a large set of process data has enable us to obtain a more reliable and rapid methodology of capital-cost forecasting. All that needs to be used is a basic flow-sheet indicating : the nature of the products to be treated, both the functional modules and their operational conditions and a material balance detailing the main streams. Results will be presented by monographs or by a set of equations, geard for pocket-calculator programs. Devised for use mainly in the early stages of process development, where a detailed equipment configuration is not available, the method has nevertheless shown good accuracy for detailed process estimating
Rousseau, Vanessa. "Modèles fragilisés de survie relative appliqués aux données censurées dans le cas de la transplantation rénale." Montpellier 1, 2008. http://www.theses.fr/2008MON1T012.
Full textSalzenstein, Fabien. "Modele markovien flou et segmentation statistique non suupervisee d'images." Rennes 1, 1996. http://www.theses.fr/1996REN10194.
Full textCapderou, Sami. "Estimation statistique non paramétrique appliquée à la surveillance des eaux côtières." Thesis, Bordeaux, 2018. http://www.theses.fr/2018BORD0128/document.
Full textThe protection of the environment, in particular aquatic systems, should be tackled as one of the top priorities of our society. The use of biological sensors to continuously test water quality is a possible way of monitoring aquatic environments. This approach has been successfully implemented on bivalve molluscs equipped with light electrodes that respect their natural behaviour, we then speak of valvometry. The purpose of this thesis is to automatically calculate and process the velocity of movement of bivalve mollusc valves in various aquatic environments. Years of recordings already acquired will allow us, from our models, to detect if there are variations in the speed of movement of the valves related to temperature variations. In particular, we studied the derivatives of different non parametric estimators of the regression function : the recursive Nadaraya-Watson estimator, the Johnston estimator, the Wand-Jones estimator and the Révész estimator. We also considered the deterministic version of the Nadaraya-Watson estimator. For each of the estimators we conducted a study on the asymptotic behaviour especially on the almost sure convergence and th asymptotic normality. We digitally illustrated these properties and applied these new estimation methods to real data to validate, or not, the environmental assumptions made by biologists
PEYRE, JEAN-LOUIS. "Statistiques d'ordre eleve appliquees a l'identification de signaux analogiques stationnaires a bande limitee." Nice, 1996. http://www.theses.fr/1996NICE4952.
Full textFortier, Susie. "Structure d'association pour une statistique morphologique appliquée à une image de Voronoï." Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2000. http://www.collectionscanada.ca/obj/s4/f2/dsk2/ftp03/MQ67270.pdf.
Full textCrispel, Simon. "Modélisation statistique appliquée aux écoulements dispersés laminaires : mise en oeuvre et simulation." Grenoble INPG, 2002. http://www.theses.fr/2002INPG0029.
Full textVan, den Broeck Samuel. "Optique statistique appliquée à la granulométrie submicronique : simulation d'un signal gaussien lorentzien." Rouen, 1998. http://www.theses.fr/1998ROUES020.
Full textDuchamps, Jean-Marc. "Apport des techniques statistiques pour l'exploitation des diagraphies instantanées en génie civil." Bordeaux 1, 1988. http://www.theses.fr/1988BOR1A635.
Full textLétang, Jean-Michel. "Intégration temporelle et régularisation statistique appliquées à la détection d'objets mobiles dans une séquence d'images." Grenoble INPG, 1993. http://www.theses.fr/1993INPG0149.
Full textNouvellet, Adrien. "Avancées récentes en traitement statistique du signal appliquées à l'estimation et la détection d'ondes infrasonores." Thesis, Paris, ENST, 2016. http://www.theses.fr/2016ENST0011/document.
Full textThe core of this thesis is the infrasound signal processing and particularly the estimation and detection using a sensor array. The themes presented here are well-suited to experimentation and we tried, as much as possible, to illustrate the theoretical results with real data. The first part describes the implications and difficulties of infrasonic detection. We review the state-of-the-art of detection techniques based on hypothesis testing and supplement them with the aim of automatic detection. The second part of this thesis highlights the error on the angle of arrival estimation due to considering the arrays as planar (common approximation). In response to this problem, a new estimator considering the full geometry of the array and including an a priori on the speed of the infrasonic wave is derived and studied. We show that the described estimator significantly improve performance. The third part introduces the modeling of infrasonic signal as autoregressive process. This modeling allows us to derived a sequential approach to detect infrasound based on the beamforming and the detection of abrupt changes. The last part of this manuscript aims to propose an alternative "signal'' model. We present new results highlighting a phenomenon of loss of coherence of the signals recorded by different sensors. This work leads to the generation of synthetic infrasonic signal in order to study the performance of the detection algorithms
Thibaudeau, Christian. "Développement d'algorithmes de reconstruction statistique appliqués en tomographie rayons-X assistée par ordinateur." Mémoire, Université de Sherbrooke, 2010. http://savoirs.usherbrooke.ca/handle/11143/1539.
Full textDécoste, Anne-Marie. "Méthodes non paramétriques appliquées à l'échantillonnage optimal à réponses partielles." Sherbrooke : Université de Sherbrooke, 1999.
Find full textRose, Anaïg Le Néel Tanguy. "Appliquer la maîtrise statistique des procédés à la fabrication de comprimés pelliculés." [S.l.] : [s.n.], 2005. http://theses.univ-nantes.fr/thesemed/PHrose.pdf.
Full textCloutier, Jacinthe. "Adéquation du modèle de socialisation à la consommation appliquée au domaine de l'épargne et invariance selon le mode de collecte de données." Thesis, Université Laval, 2011. http://www.theses.ulaval.ca/2011/28438/28438.pdf.
Full textVan, Der Baan Mirko. "Deux méthodes d'interférence statistique appliquées aux données de sismique réflexion profonde : détection de signaux et localisation d'onde." Grenoble 1, 1999. https://tel.archives-ouvertes.fr/tel-00745500.
Full textMallet, Cécile. "Modélisation statistique appliquée à la propagation atmosphérique des ondes électromagnétiques et à l'observation des précipitations." Habilitation à diriger des recherches, Université de Versailles-Saint Quentin en Yvelines, 2009. http://tel.archives-ouvertes.fr/tel-00672039.
Full textGeorges, Karine. "Méthode de traitement statistique appliquée à l'exoscopie des quartz : Environnements fluviaux et littoraux de Provence." Aix-Marseille 1, 2004. http://www.theses.fr/2004AIX10040.
Full textLefieux, Vincent Carbon Michel Delacroix Michel. "Modèles semi-paramétriques appliqués à la prévision des séries temporelles cas de la consommation d'électricité /." Rennes : Université Rennes 2, 2007. http://tel.archives-ouvertes.fr/tel-00179866/fr.
Full textLepage, Maude. "La corrélation appliquée dans un contexte bayésien." Thesis, Université Laval, 2010. http://www.theses.ulaval.ca/2010/27717/27717.pdf.
Full textCOURILLEAU, EMMANUEL. "Analyse statistique de donnees routieres appliquee au developpement de modele de gestion de l'entretien." Clermont-Ferrand 2, 1997. http://www.theses.fr/1997CLF21934.
Full textBérubé, Valérie. "Modèles avancés en régression appliqués à la tarification IARD." Thesis, Université Laval, 2007. http://www.theses.ulaval.ca/2007/24329/24329.pdf.
Full textLefieux, Vincent. "Modèles semi-paramétriques appliqués à la prévision des séries temporelles : cas de la consommation d’électricité." Phd thesis, Rennes 2, 2007. https://theses.hal.science/tel-00179866/fr/.
Full textRéseau de Transport d’Electricité (RTE), in charge of operating the French electric transportation grid, needs an accurate forecast of the power consumption in order to operate it correctly. The forecasts used everyday result from a model combining a nonlinear parametric regression and a SARIMA model. In order to obtain an adaptive forecasting model, nonparametric forecasting methods have already been tested without real success. In particular, it is known that a nonparametric predictor behaves badly with a great number of explanatory variables, what is commonly called the curse of dimensionality. Recently, semiparametric methods which improve the pure nonparametric approach have been proposed to estimate a regression function. Based on the concept of ”dimension reduction”, one those methods (called MAVE : Moving Average -conditional- Variance Estimate) can apply to time series. We study empirically its effectiveness to predict the future values of an autoregressive time series. We then adapt this method, from a practical point of view, to forecast power consumption. We propose a partially linear semiparametric model, based on the MAVE method, which allows to take into account simultaneously the autoregressive aspect of the problem and the exogenous variables. The proposed estimation procedure is practicaly efficient
Pons, Isabelle. "Méthodes de segmentation bayésienne appliquées aux images SAR : théorie et mise en oeuvre." Nice, 1994. http://www.theses.fr/1994NICE4714.
Full textGrelaud, Aude. "Méthodes sans vraisemblance appliquées à l'étude de la sélection naturelle et à la prédiction de structure tridimensionnelle des protéines." Paris 9, 2009. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=2009PA090048.
Full text