Dissertations / Theses on the topic 'Estimation de fonction'
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 'Estimation de fonction.'
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.
Gardes, Laurent. "Estimation d'une fonction quantile extrême." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2003. http://tel.archives-ouvertes.fr/tel-00005185.
Full textAKOUCHE, MAHIEDINE. "Estimation de la fonction caracteristique." Paris 6, 1993. http://www.theses.fr/1993PA066005.
Full textBirichinaga, Edurne. "Estimation spline de la moyenne d'une fonction aléatoire." Pau, 1987. http://www.theses.fr/1987PAUU3021.
Full textBirichinaga, Edurne. "Estimation spline de la moyenne d'une fonction aléatoire." Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37603023r.
Full textTurcotte, Jean-Philippe. "Estimation par densités prédictives." Mémoire, Université de Sherbrooke, 2013. http://hdl.handle.net/11143/6606.
Full textPoulin, Nicolas. "Estimation de la fonction des quantiles pour des données tronquées." Littoral, 2006. http://www.theses.fr/2006DUNK0159.
Full textIn the left-truncation model, the pair of random variables Y and T with respective distribution function F and G are observed only if Y ≥ T. Let (Yi,Ti) ; 1 ≤ i ≤ n be an observed sample of this pair of random variables. The quantile function of F is estimated by the quantile function of the Lynden-Bell (1971) estimator. After giving some results of the literature in the case of independant data, we consider the α-mixing framework. We obtain strong consistency with rates, give a strong representation for the estimator of the quantile as a mean of random variables with a neglible rest and asymptotic normality. As regards the second topic of this thesis, we consider a multidimensionnal explanatory random variable X of Y which plays the role of a response. We establish strong consitency and asymptotic normality of the conditional distribution function and those of the conditional quantile function of Y given X when Y is subject to truncation. Simulations are drawn to illustrate the results for finite samples
Degerine, Serge. "Fonction d'autocorrélation partielle et estimation autorégressive dans le domaine temporel." Grenoble 1, 1988. http://tel.archives-ouvertes.fr/tel-00243761.
Full textDegerine, Serge. "Fonction d'autocorrélation partielle et estimation autorégressive dans le domaine temporel." Grenoble 2 : ANRT, 1988. http://catalogue.bnf.fr/ark:/12148/cb37613056x.
Full textDegerine, Serge Le Breton Alain Van Cutsem Bernard. "Fonction d'autocorrélation partielle et estimation autorégressive dans le domaine temporel." S.l. : Université Grenoble 1, 2008. http://tel.archives-ouvertes.fr/tel-00243761.
Full textNembé, Jocelyn. "Estimation de la fonction d'intensité d'un processus ponctuel par complexité minimale." Phd thesis, Université Joseph Fourier (Grenoble), 1996. http://tel.archives-ouvertes.fr/tel-00346118.
Full textAttouch, Mohammed Kadi. "Estimation robuste de la fonction de régression pour des variables fonctionnelles." Littoral, 2009. http://www.theses.fr/2009DUNK0227.
Full textThe robust regression is an analysis of regression with capacity to be relatively insensitive to the large deviations due to some outliers observations. Within this framework, one proposes in this thesis studied the robust estimate of the function of regression, if the observations are at the same time independent, strongly mixing and the covariate is functional. Initially, on considers a succession of identically distributed independent observations. In this context, we establish the asymptotic normality of a robust family of estimators based on the kernel method. With title illustrative, our result is applied to the discrimination of the curves, the forecast time series, and to the construction of a confidence interval. In the second time, we suppose that the observations are strongly mixing, and we establish the rate of specific almost complete convergence and uniform of this family of estimators as well as asymptotic normality. Let us note, that the axes structural of the subject, namely “dimensionality” and the correlation of the observations, “dimensionality” and the robustness of the model, are well exploited in this study. Moreover, the property of the concentration of the measure of probability of the functional variable in small balls is used, this measure of concentration allows under some assumptions to propose an original solution to the problem of the curse of dimensionality and thus to generalize the results already obtaines in the multivariate framework. To illustrate the extension and the contribution of our work, we show in some examples how our results can be applied to the nonstandard problems of the non-parametric statistics such as the forecast of functional time series. Our methods are applied to real data such as the economy and astronomy
Nembé, Jocelyn Grègoire Gérard. "Estimation de la fonction d'intensité d'un processus ponctuel par complexité minimale." S.l. : Université Grenoble 1, 2008. http://tel.archives-ouvertes.fr/tel-00346118.
Full textFerrigno, Sandie. "Un test d'adéquation global pour la fonction de répartition conditionnelle." Montpellier 2, 2004. http://www.theses.fr/2004MON20110.
Full textBahamonde, Natalia. "Estimation de séries chronologiques avec données manquantes." Paris 11, 2007. http://www.theses.fr/2007PA112115.
Full textPambo, Bello Kowir. "Estimation à noyau de la fonction de hasard pour des variables censurées." Versailles-St Quentin en Yvelines, 2011. http://www.theses.fr/2011VERS0030.
Full textThis thesis deals with kernel estimates of the hazard function (denoted by λ) in the case when the variables are censored. It composed of three chapters. In each of the first two chapters, we construct a kernel estimator by using a double recursive stochastic algorithm (two-time-scale stochastic algorithm), and then establish its convergence in law. We compare these estimates with a nonrecursive kernel estimator. We show that the asymptotic rate of the recursive estimator λn and the one of the nonrecursive version are similar. However, in terms of confidence interval estimation, we show that it is better to use the estimator λn rather than the nonrecursive version: for the same asymptotic level, the width of the interval obtained by the recursive estimator is smaller than the one obtained with the nonrecursive version. In the third chapter, we first recall the notions of large deviations and moderate deviations, then we establish the pointwise and uniform moderate deviations principles for the sequence (eλ̃n-λ) where eλ̃n is a nonrecursive estimator
Mint, El Mouvid Mariem. "Sur l'estimateur linéaire local de la fonction de répartition conditionnelle." Montpellier 2, 2000. http://www.theses.fr/2000MON20162.
Full textBelalia, Mohamed. "Estimation et inférence non paramétriques basées sur les polynômes de Bernstein." Thèse, Université de Sherbrooke, 2016. http://hdl.handle.net/11143/8558.
Full textOuassou, Idir. "Estimation sous contraintes pour des lois à symétrie sphérique." Rouen, 1999. http://www.theses.fr/1999ROUES031.
Full textMahiou, Ramdane. "Sur l'estimation d'une fonction-frontière et l'enveloppe convexe d'un échantillon." Paris 6, 1988. http://www.theses.fr/1988PA066379.
Full textFERRIGNO, Sandie. "Un test d'adéquation global pour la fonction de répartition conditionnelle." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2004. http://tel.archives-ouvertes.fr/tel-00008559.
Full textl'on doit valider pour justifier son utilisation. Dans ce travail, on propose une approche globale où toutes les hypothèses faites pour asseoir ce modèle sont testées simultanément.
Plus précisément, on construit un test basé sur une quantité qui permet de canaliser toute l'information liant X à Y : la fonction de répartition conditionnelle de Y sachant (X = x) définie par F(y|x)=P(Y<=y|X=x). Notre test compare la valeur prise par l'estimateur polynômial local de F(y|x) à une estimation paramétrique du modèle supposé et rejette sa
validité si la distance entre ces deux quantités est trop grande. Dans un premier temps, on considère le cas où la fonction de répartition supposée est entièrement spécifiée et, dans
ce contexte, on établit le comportement asymptotique du test. Dans la deuxième partie du travail, on généralise ce résultat au cas plus courant en pratique où le modèle supposé contient un certain nombre de paramètres inconnus. On étudie ensuite la puissance locale du test en déterminant son comportement asymptotique local sous des suites d'hypothèses contigües. Enfin, on propose un critère de choix de la fenêtre d'ajustement qui intervient lors de l'étape d'estimation polynômiale locale de la fonction de répartition conditionnelle.
Bourgey, Mathieu. "Estimation des risques de développer la maladie coeliaque en fonction d'informations génétiques et familiales." Paris 11, 2007. http://www.theses.fr/2007PA11T034.
Full textHamrouni, Zouhir. "Inférence statistique par lissage linéaire local pour une fonction de régression présentant des dicontinuités." Université Joseph Fourier (Grenoble), 1999. http://tel.archives-ouvertes.fr/tel-00004840.
Full textSchmitz, Morgan A. "Euclid weak lensing : PSF field estimation." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS359/document.
Full textAs light propagates through the Universe, its path is altered by the presence of massive objects. This causes a distortion of the images of distant galaxies. Measuring this effect, called weak gravitational lensing, allows us to probe the large scale structure of the Universe. This makes it a powerful source of cosmological insight, and can in particular be used to study the distribution of dark matter and the nature of Dark Energy. The European Space Agency’s upcoming Euclid mission is a spaceborne telescope with weak lensing as one of its primary science objectives.In practice, the weak lensing signal is recovered from the measurement of the shapes of galaxies. The images obtained by any optical instrument are altered by its Point Spread Function (PSF), caused by various effects: diffraction, imperfect optics, atmospheric turbulence (for ground-based telescopes)… Since the PSF also alters galaxy shapes, it is crucial to correct for it when performing weak lensing measurements. This, in turn, requires precise knowledge of the PSF itself.The PSF varies depending on the position of objects within the instrument’s focal plane. Unresolved stars in the field provide a measurement of the PSF at given positions, from which a PSF model can be built. In the case of Euclid, star images will suffer from undersampling. The PSF model will thus need to perform a super-resolution step. In addition, because of the very wide band of its visible instrument, variations of the PSF with the wavelength of incoming light will also need to be accounted for.The main contribution of this thesis is the building of novel PSF modelling approaches. These rely on sparsity and numerical optimal transport. The latter enables us to propose the first method capable of building a polychromatic PSF model, using no information other than undersampled star images, their position and spectra. We also study the propagation of errors in the PSF to the measurement of galaxy shapes
Damon, Pierre-Marie. "Estimation pour le développement de systèmes d'aide à la conduite des véhicules à deux-roues motorisés." Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLE034/document.
Full textThe road accident investigations are unanimous: Powered Two-Wheeled Vehicles (P2WV) users are the most vulnerable on the road. In France, for a travelled kilometer, a P2WV rider has 24 times more risk to be killed than a car driver. In addition to being "naked" vehicles highly exposed to dangers, P2WV are naturally unstable with a precarious stability. That is why vehicle loss of control is clearly identified as a recurring problem into accident causation. By analogy with the driving assistance systems developed for cars (ABS, ESP, etc.), most of these accidents could be avoided with solutions dedicated to P2WV.This PhD takes place in this context and proposed original estimation algorithms. Indeed, some P2WV dynamic information, necessary for detecting risk situations, are not measurable or they require the use of costly sensors. For both technical or economic reasons, the use of observation techniques turns out to be an adequate solution in the development of P2WV riding assistance systems. Such techniques allow to estimate the vehicle dynamics while reducing the number of sensors and overcoming the unmeasurability of some dynamics states.A first part of my work is dedicated to model-based observers. In this context, an unknown input observer, a non-linear Luenberger observer and an algebraic observer were proposed. Whereas, the second part deals with the estimation based on visual perception techniques. In this last, a first algorithm was proposed to estimate the P2WV position on the road while predicting the road geometry. Then, an extension of this work was developed to reconstruct the P2WV roll angle using only images. Finally, a vision-based risk function was studied to characterize the P2WV steering behavior.All along this manuscript, the effectiveness of the proposed solutions were demonstrated through validations with the advanced simulator BikeSim framework or on experimental data
Jouvie, Camille. "Estimation de la fonction d'entrée en tomographie par émission de positons dynamique : application au fluorodesoxyglucose." Phd thesis, Université Paris Sud - Paris XI, 2013. http://tel.archives-ouvertes.fr/tel-00966453.
Full textJouvie, Camille. "Estimation de la fonction d’entrée en tomographie par émission de positons dynamique : application au fluorodesoxyglucose." Thesis, Paris 11, 2013. http://www.theses.fr/2013PA112303/document.
Full textPositron Emission Tomography (PET) is a method of functional imaging, used in particular for drug development and tumor imaging. In PET, the estimation of the arterial plasmatic activity concentration of the non-metabolized compound (the "input function") is necessary for the extraction of the pharmacokinetic parameters. These parameters enable the quantification of the compound dynamics in the tissues. This PhD thesis contributes to the study of the input function by the development of a minimally invasive method to estimate the input function. This method uses the PET image and a few blood samples. In this work, the example of the FDG tracer is chosen. The proposed method relies on compartmental modeling: it deconvoluates the three-compartment-model. The originality of the method consists in using a large number of regions of interest (ROIs), a large number of sets of three ROIs, and an iterative process. To validate the method, simulations of PET images of increasing complexity have been performed, from a simple image simulated with an analytic simulator to a complex image simulated with a Monte-Carlo simulator. After simulation of the acquisition, reconstruction and corrections, the images were segmented (through segmentation of an IRM image and registration between PET and IRM images) and corrected for partial volume effect by a variant of Rousset’s method, to obtain the kinetics in the ROIs, which are the input data of the estimation method. The evaluation of the method on simulated and real data is presented, as well as a study of the method robustness to different error sources, for example in the segmentation, in the registration or in the activity of the used blood samples
Mohamed, Zain-Eddine. "Étude économétrique de la fonction d'investissement : estimation d'un modèle à plusieurs régimes sur données sectorielles." Toulouse 1, 1987. http://www.theses.fr/1987TOU10027.
Full textInvestment decisions concern four maskets (product market, capital market, financial market, labot market). Desequilibrium in these markets implicate the investment demand. One can distinguish the walrassien demand model, and the effective demand models which incorporate different constraints from certain markets. Our object is to distinguish the intensity of that constraints by the estimation of a switching model. Our framework is an economy a little disaggregated, here the different branches of the manufacture in france
Mohamed, Zain-Eddine. "Etude économétrique de la fonction d'investissement estimation d'un modèle à plusieurs régimes sur données sectorielles /." Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37608115m.
Full textDiack, Cheikh Ahmed Tidiane. "Test de convexité pour une fonction de régression." Toulouse 3, 1997. http://www.theses.fr/1997TOU30165.
Full textCorbier, Christophe. "Contribution à l’estimation robuste de modèles dynamiques : Application à la commande de systèmes dynamiques complexes." Thesis, Paris, ENSAM, 2012. http://www.theses.fr/2012ENAM0041/document.
Full textComplex dynamic systems identification remains a concern when prediction errors contain innovation outliers. They have the effect to damage the estimated model if the estimation criterion is badly chosen and badly adapted. The consequence is the contamination of the distribution of these errors; this distribution presents heavy tails and deviates of the normal distribution. To solve this problem, there is a robust estimator's class, less sensitive to the outliers, which treat the transition between residuals of very different levels in a softer way. The Huber's M-estimators belong to this class. They are associated to a mixed L2 - L1 norm, related to a disturbed Gaussian distribution model, namely gross error model. From this formal context, in this thesis we propose a set of estimation and validation tools of black-box linear and pseudo-linear models, with extension of the noise interval to low values of the tuning constant in the Huber's norm. We present the convergence properties of the robust estimation criterion and the robust estimator. We show that the extension of the noise interval reduces the sensitivity of the bias of the estimator and improves the robustness to the leverage points. Moreover, for a pseudo-linear model structure, we present a new context, named L-FTE, with a new method to determine L, in order to linearize the gradient and the Hessien of estimation criterion and the asymptotic covariance matrix of the estimator. From these expressions, a robust version of the FPE validation criterion is established and we propose a new decisional tool for the estimated model choice. Experiments on simulated and real systems are presented and analyzed
Bouatou, Mohamed. "Estimation non linéaire par ondelettes : régression et survie." Phd thesis, Université Joseph Fourier (Grenoble), 1997. http://tel.archives-ouvertes.fr/tel-00004921.
Full textMarjani, Mohammed. "Interpolation linéaire associée à un générateur de semi-groupe intégré et estimation de la fonction K." Besançon, 1990. http://www.theses.fr/1990BESA2014.
Full textKlein, Etienne. "Estimation de la fonction de dispersion du pollen. Application a la dissemination de transgenes dans l'environnement." Paris 11, 2000. http://www.theses.fr/2000PA112027.
Full textGrimaud, Agnès. "Modélisation et estimation de la dispersion du pollen de maïs : estimation dans des modèles à volatilité stochastique." Paris 7, 2005. https://tel.archives-ouvertes.fr/tel-00011584.
Full textCarpentier, Florence. "Modélisations de la dispersion du pollen et estimation à partir de marqueurs génétiques." Thesis, Montpellier 2, 2010. http://www.theses.fr/2010MON20101.
Full textPollen dispersal is a major component of gene flow in plants. It determines to genetic diversity and spatial genetic structure.Studying it at the scale of a single reproduction event enables to understand the impact of current changes (fragmentation, anthropization ...) and to propose conservation practices.Two types of methods, based on microsatellite markers, estimate pollen dispersal functions : (i) direct methods (e.g. mating model) based on paternity assignment require exhaustif sampling (position and genotype of individuals in the study plot, genotypes of seeds harvested on mothers); (ii) indirect methods (e.g. TwoGener), require a weaker sampling (seeds genotypes, genotypes and positions of their mothers) and summarize data through genetic indices.We propose a statistical formalization of both types of methods and show that they rely on different dispersal functions : the direct methods estimate a potential forward function (pollen transfer from the father), whereas the indirect methods estimate an integrative backward one (from fecondation to father existence). We exhibit the link between forward and backward functions, assumptions leading to their equivalence and constrains affecting the backward functions.Finally, we develop an Approximate Bayesian Computation method, which enable (i) a forward estimation, (ii) with credibility intervals, (iii) from a non exhaustive dataset and partial information (e.g. positions without genotypes) and (iv) the use of different dispersal models
Muñiz, Alvarez Lilian. "Estimation nonparamétrique de la structure de covariance des processus stochastiques." Toulouse 3, 2010. http://thesesups.ups-tlse.fr/1089/.
Full textThe main objective of this thesis is the development of nonparametric methods for estimating the covariance of a stochastic process. Assuming different conditions on the process, estimators of the covariance function are introduced, having the property of being non-negative definite functions. In addition, a method for estimating the covariance matrix of a stochastic process in a high dimensional setting is proposed. Our work is organized as follows: in Chapter 1 we give a general introduction, where we present briefly the concepts and definitions underlying our work. Then come three chapters, detailing the proposed new estimation methods. More precisely, in each chapter we have developed different nonparametric estimation techniques: function approximation by wavelet thresholding in Chapter 2, model selection in Chapter 3, and estimation by Group-Lasso penalization in Chapter 4. The theoretical behavior of the estimators is studied in all cases and its good practical performances are shown in some numerical examples
Borla, Andreea. "Estimation non paramétrique de la dérivée fractionnaire de la fonction de répartition : avec une application statistique aux tests d'ajustement." Aix-Marseille 2, 2010. http://www.theses.fr/2010AIX24005.
Full textWe propose an estimator for the (a) fractional derivative of a distribution function. The estimator is based on finite differences of the empirical distribution function. The asymptotic bias, variance and consistency of the estimator are studied. It depends on a "smoothing parameter" whose behavior is similar to the bandwidth of a kernel estimator. Given that we note that an endogenous truncation can be defined in a natural way, we discuss the implementation of a truncated version of the estimator. This will allow a sharper analysis of the asymptotic behaviour of the estimator with respect to the sample size and the order of the fractional derivative. In the third chapter we will introduce another estimator for the CDF, a smoother one, the kernel estimator, and thus a second smothing factor. A natural endogenous truncation to the right ensures a tractacle definition in practice and allows deducing the asymptotic properties of the estimator. The optimal choice of the smoothing parameters is studied. Finally, we propose a goodness-of-fit test. Using simulations, we will show that there is an optimal order of differentiation that maximizes the power of the test, which depends on the alternative. Consequently test based on the integer order derivatives are not necessarily the one with the highest power. The optimum order of differentiation changes depending on the parameters of the alternative distribution and the sample size
NEGRI, ILIA. "Efficacite globale de la fonction de repartition empirique dans le cas d'un processus de diffusion ergodique." Le Mans, 1998. http://www.theses.fr/1998LEMA1007.
Full textGeffray, Ségolen. "Estimation non-paramétrique de données censurées dans un cadre multi-états." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2006. http://tel.archives-ouvertes.fr/tel-00138280.
Full textrécurrents.
Dans le cadre des risques concurrents, on s'intéresse aux fonctions
d'incidences cumulées : elles correspondent à la probabilité qu'un évènement d'un certain type se
produise avant un instant donné. Ces fonctions sont estimées de façon non-paramétrique au moyen
de l'estimateur de Aalen-Johansen. Des résultats d'approximation forte, de loi du logarithme
itéré et de convergence faible pour des processus basés sur l'estimateur de Aalen-Johansen sont
établis. Des bandes de confiance sont construites et simulées. Une extension du modèle de
Koziol-Green est aussi considérée.
Dans le cadre d'évènements récurrents, des fonctions d'incidences cumulées conditionnelles sont
estimées de façon non-paramétrique. Les estimateurs proposés sont consistants et leur
comportement à distance finie est illustré sur des données réelles et simulées.
Bouhadjera, Feriel. "Estimation non paramétrique de la fonction de régression pour des données censurées : méthodes locale linéaire et erreur relative." Thesis, Littoral, 2020. http://www.theses.fr/2020DUNK0561.
Full textIn this thesis, we are interested in developing robust and efficient methods in the nonparametric estimation of the regression function. The model considered here is the right-hand randomly censored model which is the most used in different practical fields. First, we propose a new estimator of the regression function by the local linear method. We study its almost uniform convergence with rate. We improve the order of the bias term. Finally, we compare its performance with that of the classical kernel regression estimator using simulations. In the second step, we consider the regression function estimator, based on theminimization of the mean relative square error (called : relative regression estimator). We establish the uniform almost sure consistency with rate of the estimator defined for independent and identically distributed observations. We prove its asymptotic normality and give the explicit expression of the variance term. We conduct a simulation study to confirm our theoretical results. Finally, we have applied our estimator on real data. Then, we study the almost sure uniform convergence (on a compact set) with rate of the relative regression estimator for observations that are subject to a dependency structure of α-mixing type. A simulation study shows the good behaviour of the studied estimator. Predictions on generated data are carried out to illustrate the robustness of our estimator. Finally, we establish the asymptotic normality of the relative regression function estimator for α-mixing data. We construct the confidence intervals and perform a simulation study to validate our theoretical results. In addition to the analysis of the censored data, the common thread of this modest contribution is the proposal of two alternative prediction methods to classical regression. The first approach corrects the border effects created by classical kernel estimators and reduces the bias term. While the second is more robust and less affected by the presence of outliers in the sample
Geffray, Segolen. "Estimation non-paramétrique de données censurées dans un cadre multi-états." Paris 6, 2006. http://www.theses.fr/2006PA066264.
Full textMéziani, Katia. "Estimations et tests non paramétriques en tomographie quantique homodyne." Phd thesis, Université Paris-Diderot - Paris VII, 2008. http://tel.archives-ouvertes.fr/tel-00351294.
Full textSt, Fleur Sadrac. "Estimation des mouvements sismiques à Port-au-Prince (Haïti) : mesures des amplifications locales et simulations numériques." Thesis, Université Côte d'Azur (ComUE), 2016. http://www.theses.fr/2016AZUR4099/document.
Full textIn order to help estimating the seismic ground motion expected in the Port-au-Prince area (Haiti), we characterize local site effects, pointing out the seismic waves trapped in the loose layer of Cul-de-Sac basin, and provide realistic synthetic accelerograms for an hypothetical future earthquake.To this end, we propose to analyze signals from 78 earthquakes that occurred between March 2010and February 2013, by applying two methods of spectral ratios : The H/V earthquake method and the classical spectral ratio (SSR). A strong spatial variability was observed in the measured amplifications, which is quite consistent with the heterogeneous surface geology of the area. We notice in particular strong amplification on marine sediments close to the coast. In the foothills of the Massif de la Selle the reflection of the seismic waves lead to the concentration of the wave fields that strongly amplify seismic ground motion at the top of the hills. In addition, an increase of the signal duration due to the presence of surface waves was also highlighted on some stations of the plain. For the generation of synthetic accelerograms, we first use Empirical Green functions (EGF) method. The results show that the strongest acceleration is expected in Quaternary sediments near the coast and on the ridge of south hills of Port-au-Prince. Then, a hybrid simulation method combining complex transfer functions (amplitude and phase) and the EGF simulation on bedrock was set up and validated from testing on instrumented sites in the metropolitan area of Port-au-Prince
Alarcon, Flora. "Estimation des risques de maladies dues à des mutations génétique à partir de données familiales." Phd thesis, Université Paris Sud - Paris XI, 2009. http://tel.archives-ouvertes.fr/tel-00765543.
Full textDiop, Khadim. "Estimation de la fiabilité d'un palier fluide." Thesis, Angers, 2015. http://www.theses.fr/2015ANGE0029/document.
Full textThese research is a contribution to the development of reliability theory in fluid mechanics. For the machines design and complex mechatronic systems, many fluid components are used. These components have static and dynamic sensitive characteristics and thus have a great significance on the reliabilityand lifetime of the machines and systems. Development performed focuses specifically on the reliability evaluation of a fluid bearing using a"fluid mechanics - reliability" interaction approach. This coupling requires a specific definition of the limit state function for estimating the failure probability of a fluid bearing. The Reynolds equation permits to determine the fluid bearing load capacity according to the operating conditions. Several simple geometries of fluid bearings were modeled analytically and their failure probabilities were estimated using the approximation methods FORM / SORM (First Order Reliability Method,Second Order Reliability Method) and Monte Carlo simulation
Lemaire, Vincent. "Estimation récursive de la mesure invariante d'un processus de diffusion." Phd thesis, Université de Marne la Vallée, 2005. http://tel.archives-ouvertes.fr/tel-00011281.
Full textLa principale hypothèse sur ces solutions (diffusions) est l'existence d'une fonction de Lyapounov garantissant une condition de stabilité. Par le théorème ergodique on sait que les mesures empiriques de la diffusion convergent vers une mesure invariante. Nous étudions une convergence similaire lorsque la diffusion est discrétisée par un schéma d'Euler de pas décroissant. Nous prouvons que les mesures empiriques pondérées de ce schéma convergent vers la mesure invariante de la diffusion, et qu'il est possible d'intégrer des fonctions exponentielles lorsque le coefficient de diffusion est suffisamment petit. De plus, pour une classe de diffusions plus restreinte, nous prouvons la convergence presque sûre et dans Lp du schéma d'Euler vers la diffusion.
Nous obtenons des vitesses de convergence pour les mesures empiriques pondérées et donnons les paramètres permettant une vitesse optimale. Nous finissons l'étude de ce schéma lorsqu'il y a présence de multiples mesures invariantes. Cette étude se fait en dimension 1, et nous permet de mettre en évidence un lien entre classification de Feller et fonctions de Lyapounov.
Dans la dernière partie, nous exposons un nouvel algorithme adaptatif permettant de considérer des problèmes plus généraux tels que les systèmes Hamiltoniens ou les systèmes monotones. Il s'agit de considérer les mesures empiriques d'un schéma d'Euler construit à partir d'une suite de pas aléatoires adaptés dominée par une suite décroissant vers 0.
El, Heda Khadijetou. "Choix optimal du paramètre de lissage dans l'estimation non paramétrique de la fonction de densité pour des processus stationnaires à temps continu." Thesis, Littoral, 2018. http://www.theses.fr/2018DUNK0484/document.
Full textThe work this thesis focuses on the choice of the smoothing parameter in the context of non-parametric estimation of the density function for stationary ergodic continuous time processes. The accuracy of the estimation depends greatly on the choice of this parameter. The main goal of this work is to build an automatic window selection procedure and establish asymptotic properties while considering a general dependency framework that can be easily used in practice. The manuscript is divided into three parts. The first part reviews the literature on the subject, set the state of the art and discusses our contribution in within. In the second part, we design an automatical method for selecting the smoothing parameter when the density is estimated by the Kernel method. This choice stemming from the cross-validation method is asymptotically optimal. In the third part, we establish an asymptotic properties pertaining to consistency with rate for the resulting estimate of the window-width
Garnier, Aurélie. "Dynamiques neuro-gliales locales et réseaux complexes pour l'étude de la relation entre structure et fonction cérébrales." Electronic Thesis or Diss., Paris 6, 2015. http://www.theses.fr/2015PA066562.
Full textA current issue in neuroscience is to elaborate computational models that are able to reproduce experimental data recorded with various imaging methods, and allowing us to study the relationship between structure and function in the human brain. The modeling objectives of this work are two scales and the model analysis need the development of specific theoretical and numerical tools. At the local scale, we propose a new ordinary differential equations model generating neuronal activities. We characterize and classify the behaviors the model can generate, we compare the model outputs to experimental data and we identify the dynamical structures of the neural compartment underlying the generation of pathological patterns. We then extend this approach to a new neuro-glial mass model: a bilateral coupling between the neural compartment and a new one modeling the impact of astrocytes on neurotransmitter concentrations and the feedback of these concentrations on neural activity is developed. We obtain a theoretical characterization of these feedbacks impact on neuronal excitability by formalizing the variation of a bifurcation value as a problem of optimization under constraint. Finally, we propose a network model, which node dynamics are based on the local neuro-glial mass model, embedding a neuronal coupling and a glial one. We numerically observe the differential propagations of information according to each of these coupling types and their cumulated impact, we highlight qualitatively distinct patterns of neural and glial activities of each node, and link the transitions between behaviors with the dynamical structures identified in the local models
Huynh, Quoc Vu. "ESTIMATION DES PROPRIETES POROMECANIQUES EFFECTIVES DES ARGILITES: APPORT DES METHODES D'HOMOGENEISATION." Phd thesis, Institut National Polytechnique de Lorraine - INPL, 2006. http://tel.archives-ouvertes.fr/tel-00180319.
Full textEn utilisant la solution analytique d'une inclusion sphérique isolée dans une matrice isotrope transverse infinie, l'estimation des coefficients poroélastiques effectifs des matériaux hétérogènes de type roches-composites a été réalisée.
Par la suite, on s'intéresse à la prise en compte des inclusions sphéroidales dont l'orientation ne coïncide pas avec l'axe d'orthotropie de révolution de la matrice. On propose une approche d'intégration numérique basée sur la fonction de Green. L'intégration numérique sur la sphère unité est réalisée à l'aide d'une méthode de Gauss dont la précision est discutée. L'outil numérique développé est appliqué à une roche poreuse en considérant un schéma d'homogénéisation en deux étapes et differentes fonctions de distribution d'orientation. Les résultats obtenus mettent en évidence les influences respectives des anisotropies de matrice et de l'espace poreux.
Diallo, Sara. "Interférence statistique par lissage pour la fonction de vie résiduelle moyenne présentant des discontinuités." Le Havre, 2008. http://www.theses.fr/2008LEHA0001.
Full textEstimators for location and size of a discontinuity or change-point in a smooth mean residual life model are proposed. The proposed estimators also apply to the detection of discontinuities in derivatives and therefore to the detection of change-points of slope and of higher order curvature. The proposed estimators are based on a comparison of left and right one-sided kernel smoothers. Weak convergence of a stochastic process in local differences to a Gaussian process is established for properly scaled versions of estimators for the location of a change-point. The continuous mapping theorem can then be invoked to obtain asymptotic distributions and corresponding rates of convergence for change-point estimators