Dissertations / Theses on the topic 'Estimation de la régression'
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 la régression.'
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.
Simard, Joanie. "Méthodes de régression robuste." Mémoire, Université de Sherbrooke, 2018. http://hdl.handle.net/11143/12002.
Full textGendre, Xavier. "Estimation par sélection de modèle en régression hétéroscédastique." Phd thesis, Université de Nice Sophia-Antipolis, 2009. http://tel.archives-ouvertes.fr/tel-00397608.
Full textLa première partie de cette thèse consiste dans l'étude du problème d'estimation de la moyenne et de la variance d'un vecteur gaussien à coordonnées indépendantes. Nous proposons une méthode de choix de modèle basée sur un critère de vraisemblance pénalisé. Nous validons théoriquement cette approche du point de vue non-asymptotique en prouvant des majorations de type oracle du risque de Kullback de nos estimateurs et des vitesses de convergence uniforme sur les boules de Hölder.
Un second problème que nous abordons est l'estimation de la fonction de régression dans un cadre hétéroscédastique à dépendances connues. Nous développons des procédures de sélection de modèle tant sous des hypothèses gaussiennes que sous des conditions de moment. Des inégalités oracles non-asymptotiques sont données pour nos estimateurs ainsi que des propriétés d'adaptativité. Nous appliquons en particulier ces résultats à l'estimation d'une composante dans un modèle de régression additif.
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 textGaïffas, Stéphane. "Régression non-paramétrique et information spatialement inhomogène." Paris 7, 2005. https://tel.archives-ouvertes.fr/tel-00011261.
Full textGaiffas, Stéphane. "Régression non-paramétrique et information spatialement inhomogène." Phd thesis, Université Paris-Diderot - Paris VII, 2005. http://tel.archives-ouvertes.fr/tel-00011261.
Full textdonnées bruitées spatialement inhomogènes (données dont la quantité
varie sur le domaine d'estimation). Le prototype d'étude est le modèle
de régression avec design aléatoire. Notre objectif est de comprendre
les conséquences du caractère inhomogène des données sur le problème
d'estimation dans le cadre d'étude minimax. Nous adoptons deux points
de vue : local et global. Du point de vue local, nous nous intéressons
à l'estimation de la régression en un point avec peu ou beaucoup de
données. En traduisant cette propriété par différentes hypothèses sur
le comportement local de la densité du design, nous obtenons toute une
gamme de nouvelles vitesses minimax ponctuelles, comprenant des
vitesses très lentes et des vitesses très rapides. Puis, nous
construisons une procédure adaptative en la régularité de la
régression, et nous montrons qu'elle converge avec la vitesse minimax
à laquelle s'ajoute un coût minimal pour l'adaptation locale. Du point
de vue global, nous nous intéressons à l'estimation de la régression
en perte uniforme. Nous proposons des estimateurs qui convergent avec
des vitesses dépendantes de l'espace, lesquelles rendent compte du
caractère inhomogène de l'information dans le modèle. Nous montrons
l'optimalité spatiale de ces vitesses, qui consiste en un renforcement
de la borne inférieure minimax classique pour la perte uniforme. Nous
construisons notamment un estimateur asymptotiquement exact sur une
boule de Hölder de régularité quelconque, ainsi qu'une bande de
confiance dont la largeur s'adapte à la quantité locale de données.
Cao, Yun. "Inégalités d'Oracle pour l'Estimation de la Régression." Phd thesis, Université de Provence - Aix-Marseille I, 2008. http://tel.archives-ouvertes.fr/tel-00341752.
Full textAfin d'obtenir des inégalités d'oracle, on applique l'inégalité de Doob pour le processus de Wiener pour l'estimation par polynômes ; dans le cas de l'estimation par splines, on introduit le processus ordonné en généralisant le processus de Wiener.
Blondin, David. "Lois limites uniformes et estimation non-paramétrique de la régression." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2004. http://tel.archives-ouvertes.fr/tel-00011943.
Full textDelsol, Laurent. "Régression sur variable fonctionnelle : estimation, tests de structure et applications." Phd thesis, Université Paul Sabatier - Toulouse III, 2008. http://tel.archives-ouvertes.fr/tel-00449806.
Full textDia, Galaye. "Statistiques d'ordre dans les processus ponctuels : estimation de la régression." Paris 6, 1986. http://www.theses.fr/1986PA066257.
Full textPortier, François. "Réduction de la dimension en régression." Phd thesis, Université Rennes 1, 2013. http://tel.archives-ouvertes.fr/tel-00871049.
Full textBrua, Jean-Yves. "Estimation non paramétrique pour des modèles de diffusion et de régression." Phd thesis, Université Louis Pasteur - Strasbourg I, 2008. http://tel.archives-ouvertes.fr/tel-00338286.
Full textPour un modèle de régression non paramétrique et hétéroscédastique, où l'écart-type du bruit dépend à la fois du régresseur et de la fonction de régression supposée appartenir à une classe höldérienne faible de régularité connue, nous montrons qu'un estimateur à noyau est asymptotiquement efficace. Lorsque la régularité de la fonction de régression est inconnue, nous obtenons la vitesse de convergence minimax adaptative des estimateurs sur une famille de classes höldériennes. Enfin, pour un modèle de diffusion où la dérive appartient à un voisinage höldérien faible centré en une fonction lipschitzienne, nous présentons la construction d'un estimateur à noyau asymptotiquement efficace.
Baraud, Yannick. "Sélection de modèles et estimation adaptative dans différents cadres de régression." Paris 11, 1998. http://www.theses.fr/1998PA112002.
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
Camlong-Viot, Christine. "Modèle additif de régression sous des conditions de mélange." Toulouse 3, 2000. http://www.theses.fr/2000TOU30168.
Full textSerot, Isabelle. "Temps local et estimation de régression dans les processus à temps continu." Paris 6, 2002. http://www.theses.fr/2002PA066335.
Full textFragneau, Christopher. "Estimation dans le modèle de régression monotone single index en grande dimension." Thesis, Paris 10, 2020. http://www.theses.fr/2020PA100069.
Full textThe framework of this thesis is the monotone single-index model which assumes that a real variable Y is linked to a d dimensional real vector X through the relationship E[Y|X] = f(aTX) a.s., where the real monotonic function f and aT, the transposed vector of a are unknown. This model is well-known in economics, medecine and biostatistics where the monotonicity of f appears naturally. Given n replications of (X,Y) and assuming that a belongs to S, the d unit dimensional sphere, my main aim is to estimate (a,f) in the high dimensional context, where d is allowed to depend on n and to grow to infinity with n. First Chapter introduces the theory of monotone estimation, the high-dimensional context, and the single-index model. Second Chapter studies the minimizers of least-squares and maximum likelihood population criteria over classes K of couples (b,g) where b belongs to a subset of S and g is monotonic. My results are are needed for constrained estimators convergence purposes over K, the aim of the Third Chapter. In a setting where d depends on n and the distribution of X is eitherbounded or sub-Gaussian, I establish the rates of convergence of the estimators of f(aT·), a and f in case where (a,f) ∈ K, as well as the consistency of estimators of f(aT·), otherwise. Fourth Chapter furnishes an estimation method of (a,f) when X is assumed to be a Gaussian vector. This method fits a mispecified linear model, and estimates its parameter vector thanks to the de-sparcified Lasso method of Zhang and Zhang (2014). I show that the resulting estimator divided by its Euclidean norm is Gaussian and converges to a, at parametric rate. I provide estimators of f(aT·) and f, and I establish their rates of convergence
Slaoui, Yousri. "Application des méthodes d'approximations stochastiques à l'estimation de la densité et de la régression." Phd thesis, Université de Versailles-Saint Quentin en Yvelines, 2006. http://tel.archives-ouvertes.fr/tel-00131964.
Full textSidi, Zakari Ibrahim. "Sélection de variables et régression sur les quantiles." Thesis, Lille 1, 2013. http://www.theses.fr/2013LIL10081/document.
Full textThis work is a contribution to the selection of statistical models and more specifically in the selection of variables in penalized linear quantile regression when the dimension is high. It focuses on two points in the selection process: the stability of selection and the inclusion of variables by grouping effect. As a first contribution, we propose a transition from the penalized least squares regression to quantiles regression (QR). A bootstrap approach based on frequency of selection of each variable is proposed for the construction of linear models (LM). In most cases, the QR approach provides more significant coefficients. A second contribution is to adapt some algorithms of "Random" LASSO (Least Absolute Shrinkage and Solution Operator) family in connection with the QR and to propose methods of selection stability. Examples from food security illustrate the obtained results. As part of the penalized QR in high dimension, the grouping effect property is established under weak conditions and the oracle ones. Two examples of real and simulated data illustrate the regularization paths of the proposed algorithms. The last contribution deals with variable selection for generalized linear models (GLM) using the nonconcave penalized likelihood. We propose an algorithm to maximize the penalized likelihood for a broad class of non-convex penalty functions. The convergence property of the algorithm and the oracle one of the estimator obtained after an iteration have been established. Simulations and an application to real data are also presented
Hedli-Griche, Sonia. "Estimation de l'opérateur de régression pour des données fonctionnelles et des erreurs corrélées." Université Pierre Mendès France (Grenoble), 2008. http://www.theses.fr/2008GRE21009.
Full textIn the research work that we present in this thesis, we study the problem of nonparametric modelization when the statistical data are represented by curves. More precisely, we are interested in the problems of prediction from an explanatory random variable that takes values in some, eventually, infinite dimensional space. Recently, some work has been realised in the functional operatoriel estimation under the independence assumptions of the functional data. In this thesis, we consider that the functional data are dependent and that the error process is stationary (with short or long memory). We have studied and estimated the regression operator under different set-ups: when the functional data (dependent) are deterministic or random, when the error process is a short or long memory, the asymptotic normality when the error process is negatively associated, the local/global choice of the bandwidth, the study of the relevancy of our theoretical results to simulated data and then to real data
Yao, Anne-Françoise. "Un modèle semi-paramétrique pour variables fonctionnelles : la régression inverse fonctionnelle." Toulouse 3, 2001. http://www.theses.fr/2001TOU30122.
Full textSaracco, Jérôme. "Contributions à la régression inverse par tranchage : sliced inverse regression (S.I.R.)." Toulouse 3, 1996. http://www.theses.fr/1996TOU30185.
Full textDiack, Cheikh Ahmed Tidiane. "Test de convexité pour une fonction de régression." Toulouse 3, 1997. http://www.theses.fr/1997TOU30165.
Full textCherfi, Mohamed. "Estimation par minimum de Ø-divergences." Paris 6, 2010. http://www.theses.fr/2010PA066389.
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 textColliez, Johan. "Estimation robuste du mouvement dans les séquences d'images basée sur la régression par vecteurs supports." Littoral, 2009. http://www.theses.fr/2009DUNK0231.
Full textOne of the main tasks in computer vision is to extract the relevant information from an images sequence by using the regression and model adjustment theory. However, the presence of noise and ouliers has the effect of altering the task of estimating the structure of the underlying model. Hence, the need of using robust estimators against the errors inherent to natural scenes images. In this work, we propose a new robust estimator based on support vectors machines. This estimator is a weighted version of regression by support vectors. It assigns a heterogeneous penality to observations according that they belong to inliers or outliers classes. Hard and soft penalisations were considered and an iterative approach was applied to extract the dominant structure in the data set. The many simulated sets indicate that the proposed robust estimator by support vectors has a breakdown point above 50% imroving significantly the performance of the standard regression by support vector. Moreover, it permits to extract the dominant structure in the data set with a high resistance to residual structures. The robust regression approach was applied to estimate the movement in images sequence by optic flow as well as by images matching
Vimond, Myriam. "Inférence statistique par des transformées de Fourier pour des modèles de régression semi-paramétriques." Phd thesis, Université Paul Sabatier - Toulouse III, 2007. http://tel.archives-ouvertes.fr/tel-00185102.
Full textTaupin, Marie-Luce. "Estimation semi-paramétrique pour le modèle de régression non linéaire avec erreurs sur les variables." Paris 11, 1998. http://www.theses.fr/1998PA112004.
Full textViallon, Vivian. "Processus empiriques, estimation non paramétrique et données censurées." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2006. http://tel.archives-ouvertes.fr/tel-00119260.
Full textPan, Jingjing. "Estimation des temps de retard et localisation de sources avec des systèmes Radar." Thesis, Nantes, 2018. http://www.theses.fr/2018NANT4016/document.
Full textSource localization (in far-field or in nearfield) and time-delay estimation have many practical applications. To locate a far-field source from a sensor array, only the direction of arrival (DOA) of the source is necessary. When the sources are in a nearfield situation, the wavefront of the signal is spherical and two parameters are needed to locate the sources: the direction of arrival and the distance between the source and the sensors. In this thesis, we focus on the localization of sources (both in far-field and nearfield) as well as the estimation of time-delay in the context where the signals are coherent, overlapped and with a small number of snapshots. First, we propose to combine the theory of the SVR method (support vector regression, which is a supervised learning-based regression method) with the theory of forward-backward linear prediction (FBLP). The proposed method, called FBLP-SVR, is developed for two applications: far-field source localization and time-delay estimation by using ground penetrating radar. The proposed method is evaluated by simulations and experiments. We also propose a near-field source localization method in the context where the signals are coherent and overlapped. The proposed method is based on a focusing technique, a spatial smoothing preprocessing, and a subspace method in the estimation of DOA. Then, the distances between the sources and sensors are estimated with the maximum likelihood method
Celisse, Alain. "Sélection de modèle par validation-croisée en estimation de la densité, régression et détection de ruptures." Phd thesis, Université Paris Sud - Paris XI, 2008. http://tel.archives-ouvertes.fr/tel-00346320.
Full textBouquiaux, Christel. "Semiparametric estimation for extreme values." Doctoral thesis, Universite Libre de Bruxelles, 2005. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210910.
Full textDoctorat en sciences, Orientation statistique
info:eu-repo/semantics/nonPublished
Charlier, Isabelle. "Conditional quantile estimation through optimal quantization." Thesis, Universite Libre de Bruxelles, 2015. http://www.theses.fr/2015BORD0274/document.
Full textOne of the most common applications of nonparametric techniques has been the estimation of a regression function (i.e. a conditional mean). However it is often of interest to model conditional quantiles, particularly when it is felt that the conditional mean is not representative of the impact of the covariates on the dependent variable. Moreover, the quantile regression function provides a much more comprehensive picture of the conditional distribution of a dependent variable than the conditional mean function. Originally, the “quantization” was used in signal and information theories since the fifties. Quantization was devoted to the discretization of a continuous signal by a finite number of “quantizers”. In mathematics, the problem of optimal quantization is to find the best approximation of the continuous distribution of a random variable by a discrete law with a fixed number of charged points. Firstly used for a one-dimensional signal, the method has then been developed in the multi-dimensional case and extensively used as a tool to solve problems arising in numerical probability. The goal of this thesis is to study how to apply optimal quantization in Lp-norm to conditional quantile estimation. Various cases are studied: one-dimensional or multidimensional covariate, univariate or multivariate dependent variable. The convergence of the proposed estimators is studied from a theoretical point of view. The proposed estimators were implemented and a R package, called QuantifQuantile, was developed. Numerical behavior of the estimators is evaluated through simulation studies and real data applications
Durot, Cécile. "Asymptotique fine pour l'estimateur isotonique en régression et méthodes de jackknife : applications à la comparaison de courbes de croissance." Paris 11, 1997. http://www.theses.fr/1997PA112007.
Full textDalalyan, Arnak. "Contribution à la statistique des diffusions. Estimation semiparamétrique et efficacité au second ordre.Agrégation et réduction de dimension pour le modèle de régression." Habilitation à diriger des recherches, Université Pierre et Marie Curie - Paris VI, 2007. http://tel.archives-ouvertes.fr/tel-00192080.
Full textLe premier chapitre contient une description générale des résultats obtenus en les replaçant dans un contexte historique et en présentant les motivations qui nous ont animées pour étudier ces problèmes. J'y décris également de façon informelle les idées clés des démonstrations.
Au second chapitre, je présente les définitions principales nécessaires pour énoncer de façon rigoureuse les résultats les plus importants. Ce chapitre contient également une discussion plus formelle permettant de mettre en lumière certains aspects théoriques et pratiques de nos résultats.
Usseglio-Carleve, Antoine. "Estimation de mesures de risque pour des distributions elliptiques conditionnées." Thesis, Lyon, 2018. http://www.theses.fr/2018LYSE1094/document.
Full textThis PhD thesis focuses on the estimation of some risk measures for a real random variable Y with a covariate vector X. For that purpose, we will consider that the random vector (X,Y) is elliptically distributed. In a first time, we will deal with the quantiles of Y given X=x. We thus firstly investigate a quantile regression model, widespread in the litterature, for which we get theoretical results that we discuss. Indeed, such a model has some limitations, especially when the quantile level is said extreme. Therefore, we propose another more adapted approach. Asymptotic results are given, illustrated by a simulation study and a real data example.In a second chapter, we focus on another risk measure called expectile. The structure of the chapter is essentially the same as that of the previous one. Indeed, we first use a regression model that is not adapted to extreme expectiles, for which a methodological and statistical approach is proposed. Furthermore, highlighting the link between extreme quantiles and expectiles, we realize that other extreme risk measures are closely related to extreme quantiles. We will focus on two families called Lp-quantiles and Haezendonck-Goovaerts risk measures, for which we propose extreme estimators. A simulation study is also provided. Finally, the last chapter is devoted to the case where the size of the covariate vector X is tall. By noticing that our previous estimators perform poorly in this case, we rely on some high dimensional estimation methods to propose other estimators. A simulation study gives a visual overview of their performances
Chambaz, Antoine. "Segmentation spatiale et sélection de modèle : théorie et applications statistiques." Paris 11, 2003. http://www.theses.fr/2003PA112012.
Full textWe tacke in this thesis the elaboration of an original method that provides refinement of the localization of the mobIle telecommunication traffic in urban area for France Télécom R&D. This work involves both practical and theoretical developments. Our point of view is of statistical nature. The major themes are spatial segmentation and model selection. We first introduce the various datasets from which our approach stems. They cast some light on the original problem. We motivate the choice of an heteroscedastic regression model. We then present a practical nonparametric regression method based on CART regression trees and its Bagging and Boosting extensions by resampling. The latter classical methods are designed for ho- moscedastic models. We propose an adaptation to heteroscedastic ODes, including an original analysis of variable importance. We apply the method to various traffic datasets. The final results are commented. The above practical work motivates the theoretical study of the consistency of a family of estimators of the order of a segmented model and its associated segmentation. We also cope, in a general framework of model select ion in a nested family of models, with the estimation of the order of a model. We are particularly concerned with consistency properties and rates of und er- or overestimation. We tackle the problem at stake with a linear functional approach, i. E. An approach where the events of interest are described as events concerning the empirical measute. This allows to derive general results that gather and enhance earlier ODes. A large range of techniques are involved : classical arguments of M -estimation, concentration, max- imal inequalities for dependent variables, Stein's lemma, penalization, Large and Moderate Deviations Principles for the empirical measure, à la Huber trick
Venditti, Véronique. "Aspects du principe de maximum d'entropie en modélisation statistique." Université Joseph Fourier (Grenoble), 1998. http://www.theses.fr/1998GRE10108.
Full textEzzahar, Abdessamad. "Estimation et détection d'un signal contaminé par un bruit autorégressif." Phd thesis, Grenoble 1, 1991. http://tel.archives-ouvertes.fr/tel-00339831.
Full textMartinez, Lopez Carlos Manuel. "Application de la régression en composantes principales au traitement des données acoustiques multifréquence : estimation des abondances du zooplancton." Aix-Marseille 2, 1992. http://www.theses.fr/1992AIX22104.
Full textBouhadjera, 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
Slaoui, Yousri. "Application des méthodes d'approximation stochastique à l'estimation de la densité et de la régression." Versailles-St Quentin en Yvelines, 2006. http://www.theses.fr/2006VERS0025.
Full textThe objective of this thesis is to apply the stochastic approximation methods to the estimation of a density and of a regression function. Ln the first chapter, we build up a stochastic algorithm with single stepsize, which defines a whole class of recursive kernel estimators of a probability density. We study the properties of this algorithm. We show that the two recursive kernel estimators already known correspond to two particular elements of the class of estimators defined by our stochastic algorithm : the recursive estimator introduced by Wolwerton and Wagner (1969) corresponds to a choice of stepsize of the algorithm which allows to minimize the mean squared error, while the one introduced by Duflo (1997) corresponds to a choice of stepsize which minimizes the variance. Ln the second chapter, we consider the estimator proposed by Révész (1973, 1977) to estimate a regression function r : x f---t lE [YIX = x]. His estimator rn, built up by using a single-time-scale stochastic algorithm, has a big disadvantage : the assumptions on the marginal density of X necessary to establish the convergence rate of r n are much stronger than those usually required to study the asymptotic behavior of an estimator of a regression function. We show how the application of the averaging principle of stochastic algorithms allows, by first generalizing the definition of the estimator of Révész and then by averaging this generalized estimator, to build up a recursive estimator fn which has good asymptotic properties. Ln the third chapter, we still apply stochastic approximation methods to estimate a regression function. But this time, rather than to use single¬time-scale stochastic algorithm, we show how the two-time-scale stochastic algorithms allow to build up a whole class of recursive estimators of a regression function, and we study the asymptotic properties of these estimators. This approach is much easier than the one of the second chapter : the estimators built up using the two-time-scale algorithms do not need to be averaged to have good asymptotic properties
Sauvé, Marie. "Sélection de modèles en régression non gaussienne : applications à la sélection de variables et aux tests de survie accélérés." Paris 11, 2006. http://www.theses.fr/2006PA112201.
Full textThis thesis deals with model selection in non Gaussian regression. Our aim is to get informations on a function s given only some values perturbed by noises non necessarily Gaussian. In a first part, we consider histogram models (i. E. Classes of piecewise constant functions) associated with a collection of partitions of the set on which s is defined. We determine a penalized least squares criterion which selects a partition whose associated estimator satisfies an oracle inequality. Selecting a histogram model does not always lead to an accurate estimation of s, but allows for example to detect the change-points of s. In order to perform variable selection, we also propose a non linear method which relies on the use of CART and on histogram model selection. In a second part, we consider piecewise polynomial models, whose approximation properties are better. We aim at estimating s with a piecewise polynomial whose degree can vary from region to region. We determine a penalized criterion which selects a partition and a series of degrees whose associated piecewise polynomial estimator satisfies an oracle inequality. We also apply this result to detect the change-points of a piecewise affine function. The aim of this last work is to provide an adequate stress interval for Accelerating Life Test
Brunel, Victor Emmanuel. "Non parametric estimation of convex bodies and convex polytopes." Thesis, Paris 6, 2014. http://www.theses.fr/2014PA066146/document.
Full textIn this thesis, we are interested in statistical inference on convex bodies in the Euclidean space $\R^d$. Two models are investigated. The first one consists of the observation of $n$ independent random points, with common uniform distribution on an unknown convex body. The second one is a regression model, with additive subgaussian noise, where the regression function is the indicator function of an unknown convex body. In the first model, our goal is to estimate the unknown support of the common uniform density of the observed points. In the second model, we aim either to estimate the support of the regression function, or to detect whether this support is nonempty, i.e., the regression function is nonzero. In both models, we investigate the cases when the unknown set is a convex polytope, and when we know the number of vertices. If this number is not known, we propose an adaptive method which allows us to obtain a statistical procedure performing asymptotically as well as in the case of perfect knowledge of that number. In addition, this procedure allows misspecification, i.e., provides an estimator of the unknown set, which is optimal in a minimax sense, even if the unknown set is not polytopal, in the contrary to what may have been thought. We prove a universal deviation inequality for the volume of the convex hull of the observations in the first model. We show that this inequality allows one to derive tight bounds on the moments of the missing volume of this convex hull, as well as on the moments of the number of its vertices. In the one-dimensional case, in the second model, we compute the asymptotic minimal size of the unknown set so that it can be detected by some statistical procedure, and we propose a decision rule which allows consistent testing of whether of that set is empty
Le, Thi Xuan Mai. "Estimation semi-paramétrique et application à l’évaluation de la biomasse d'anchois." Thesis, Toulouse, INSA, 2010. http://www.theses.fr/2010ISAT0006/document.
Full textThe motivation of this study is to evaluate the anchovy biomass, that is estimate the egg densities at the spawning time and the mortality rate. The data are the anchovy egg densities that are the egg weights by area unit, collected in the Gascogne bay. The problem we are faced is to estimate from these data the egg densities at the spawning time. Until now, this is done by using the classical exponential mortality model. However, such model is inadequate for the data under consideration because of the great spatial variability of the egg densities at the spawning time. They are samples of generated by a r.v whose mathematical expectation is a0 and the probability density function is fA. Therefore, we propose an extended exponential mortality model Y (tj,kj) = A (tj,kj) e-z0tj +e(tj,kj) where A(tj,kj) and e(tj,kj) are i.i.d, with the random variables A and e being assumed to be independent. Then the problem consists in estimating the mortality rate and the probability density of the random variable . We solve this semiparametric estimation problem in two steps. First, we estimate the mortality rate by fitting an exponential mortality model to averaged data. Second, we estimate the density fA by combining nonparametric estimation method with deconvolution technique and estimate the parameter z0. Theoretical results of consistence of these estimates are corroborated by simulation studies
Maatouk, Hassan. "Correspondance entre régression par processus Gaussien et splines d'interpolation sous contraintes linéaires de type inégalité. Théorie et applications." Thesis, Saint-Etienne, EMSE, 2015. http://www.theses.fr/2015EMSE0791/document.
Full textThis thesis is dedicated to interpolation problems when the numerical function is known to satisfy some properties such as positivity, monotonicity or convexity. Two methods of interpolation are studied. The first one is deterministic and is based on convex optimization in a Reproducing Kernel Hilbert Space (RKHS). The second one is a Bayesian approach based on Gaussian Process Regression (GPR) or Kriging. By using a finite linear functional decomposition, we propose to approximate the original Gaussian process by a finite-dimensional Gaussian process such that conditional simulations satisfy all the inequality constraints. As a consequence, GPR is equivalent to the simulation of a truncated Gaussian vector to a convex set. The mode or Maximum A Posteriori is defined as a Bayesian estimator and prediction intervals are quantified by simulation. Convergence of the method is proved and the correspondence between the two methods is done. This can be seen as an extension of the correspondence established by [Kimeldorf and Wahba, 1971] between Bayesian estimation on stochastic process and smoothing by splines. Finally, a real application in insurance and finance is given to estimate a term-structure curve and default probabilities
Sabbah, Camille. "Contribution à l'étude des M-estimateurs polynômes locaux." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00509898.
Full textNguyen, ThiMongNgoc. "Estimation récursive pour des modèles semi-paramétriques." Thesis, Bordeaux 1, 2010. http://www.theses.fr/2010BOR14107/document.
Full textCrambes, Christophe. "Modèles de régression linéaire pour variables explicatives fonctionnelles." Phd thesis, Université Paul Sabatier - Toulouse III, 2006. http://tel.archives-ouvertes.fr/tel-00134003.
Full textPchelintsev, 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
Dabo-Niang, Sophie. "Sur l'estimation fonctionnelle en dimension infinie : application aux diffusions." Paris 6, 2002. http://www.theses.fr/2002PA066273.
Full text