Letteratura scientifica selezionata sul tema "Profondeur de Tukey"
Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili
Indice
Consulta la lista di attuali articoli, libri, tesi, atti di convegni e altre fonti scientifiche attinenti al tema "Profondeur de Tukey".
Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.
Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.
Tesi sul tema "Profondeur de Tukey"
Cisse, Mouhamadou Moustapha. "La fonction de profondeur de Tukey". Master's thesis, Université Laval, 2019. http://hdl.handle.net/20.500.11794/34506.
Testo completoIn this memoir we define the Tukey depth function of a positive finite measure on Rd. Then we study the properties of this function, in particular the properties of continuity and convexity. We seek to establish a characterization of a measure by its depth function. That is, given μ, v finite positive measures on Rd, do we have μ = v if μ and v have the same Tukey depth function? We use the properties of the depth function to establish such a characterization when the measure satisfies certain geometric properties. Then we exhibit some approaches for computing the Tukey depth function. Finally we prove the theorem of characterisation of a discrete measure by its Tukey depth function.
Lahlou, Mimi Said. "Profondeur de Tukey et son application en contrôle de qualité multivarié /". Thèse, Trois-Rivières : Université du Québec à Trois-Rivières, 2000. http://www.uqtr.ca/biblio/notice/resume/03-2224127R.html.
Testo completoLahlou, Mimi Said. "Profondeur de Tukey et son application en contrôle de qualité multivarié". Thèse, Université du Québec à Trois-Rivières, 2000. http://depot-e.uqtr.ca/3175/1/000677395.pdf.
Testo completoGenest, Maxime. "Mesures de localisation et de dispersion et profondeur de Tukey en statistique directionnelle". Thesis, Université Laval, 2010. http://www.theses.ulaval.ca/2010/27669/27669.pdf.
Testo completoBriend, Simon. "Inference of the past of random structures and other random problems". Electronic Thesis or Diss., université Paris-Saclay, 2024. http://www.theses.fr/2024UPASM013.
Testo completoThis thesis is decomposed in three disjoint parts. The first two parts delve into dynamically growing networks. In the first part, we infer information about the past from a snapshot of the graph. We start by the problem of root finding, where the goal is to find confidence set for the root. We propose a method for uniform L-dags and analyse its performance. It is, to the best of our knowledge, the first method achieving network archaeology in general graphs. Then, we naturally extend the question of root finding to the one of seriation. Given a snapshot of a graph, is it possible to retrieve its whole ordering? We present a method and statistical guarantee of its quality in the case of uniform random recursive trees and linear preferential attachment tree. To conclude the network archaeology section, we study the root bit finding problem, where one does not try to infer the position of the root but its state. In such problems, the root is assigned a bit and is then propagated through a noisy channel during network growth. In the L-dag, we study majority voting to infer the bit of the root and we identify three different regimes depending on the noise level. In the second part of this thesis, we study the so called friendship tree, which is a random recursive tree model with complete redirection. This model display emerging properties, but unlike in the preferential attachment model they stem from a local attachment rule. We prove conjectures about degree distribution, diameter and local structure. Finally, we delve into the world of theoretical machine learning and data analysis. We study a random approximation of the Tukey depth. The Tukey depth is a powerful tool for data visualization and can be thought of as an extension of quantiles in higher dimension (they coincide in dimension 1). Its exact computation is NP-hard, and we study the performances of a classical random approximation in the case of data sets sampled from log-concave distribution