變數選取方法已經成為各領域在處理多維度資料的工具。Zhou與Hwang在2005年，為了改善James-Stein positive part估計量(JS+)只能在完全模型(full model)與原始模型(origin model)兩者去做挑選，建立了具有Minimax性質同時加上門檻值的估計量，即James-Stein with Threshoding positive part估計量(JSWT+)。由於JSWT+估計量具有門檻值，使得此估計量可以在完全模型與其線性子集下做變數選取。我們想進一步了解如果將JSWT+估計量應用於線性迴歸分析時，藉由JSWT+估計具有門檻值的性質去做變數選取的效果如何？本文目的即是利用JSWT+估計量具有門檻值的性質，建立JSWT+估計量應用於線性迴歸模型變數挑選的流程。建立模擬資料分析，以可同時做係數壓縮及變數選取的LASSO方法與我們所提出JSWT+變數選取的流程去比較係數路徑及變數選取時差異比較，最後將我們提出JSWT+變數選取的流程對實際資料攝護腺癌資料(Tibshirani，1996)做變數挑選。則當考慮解釋變數個數小於樣本個數情況下，JSWT+與LASSO在變數選取的比較結果顯示，JSWT+表現的比較好，且可直接得到估計量的理想參數。

Stoney Burns (Brent LaSalle Stein) edited and published Dallas Notes, a Dallas, Texas, underground newspaper, from November 1967 through September 1970. This thesis considers whether Burns was the unifying figure in the Dallas counterculture.

Le problème étudié dans cette thèse est celui du débruitage d'images numériques cor- rompues par un bruit blanc gaussien. Les méthodes utilisées pour récupérer une meilleure image reposent sur les patchs et sont des variantes des Non-Local Means. Les contributions de la thèse sont à la fois pratiques et théoriques. Tout d'abord, on étudie précisément l'influence des divers paramètres de la méthode. On met ensuite en lumière une lim- ite observée sur le traitement des bords par les méthodes à patchs habituelles. On donne alors une meilleure façon de combiner l'information fournie à partir des patchs pour estimer pixel par pixel. D'un point de vue théorique, on présente un cadre non asymptotique pour contrôler notre estimateur. On donne alors des résultats de type inégalités oracles pour des estimateurs vérifiant des propriétés plus restrictives. Les techniques utilisées reposent sur l'agrégation d'estimateurs, et plus particulièrement sur l'agrégation à poids exponentiels. La méthode requiert typiquement une mesure du risque, obtenue à travers un estimateur sans biais de celui-ci, par exemple par la méthode de Stein. Les méthodes de débruitage étudiées sont analysées numériquement par simulations.

Dans ce dossier je présente l'ensemble des travaux de recherche que j'ai effectués pendant les années 2007-2012. Depuis ma thèse doctorale (soutenue le 11 décembre 2006), mes travaux de recherche portent sur les risques de crédit et des applications, en particulier, sur les quatre thèmes suivants : approche de densité dans la modélisation de défaut, optimisation de portefeuille avec risques de contrepartie et de contagion, risque d'informations asymétriques sur le défaut, et méthode de Stein. L'étude de ces problèmes fait intervenir de différentes théories mathématiques, comme par exemple le grossissement de filtrations, le contrôle stochastique, les méthodes numériques probabilistes, etc., et conduit à des questions mathématiques intéressantes, souvent profondes. Parmi ces sujets, certains sont issus de motivations d'actualité financière et susceptibles d'avoir des applications dans la pratique; d'autres sont des problèmes mathématiques inspirés par la finance mais qui ont un caractère théorique.

Dans cette thèse, on s'intéresse, sous deux aspects différents, à la cohomologie des variétés de Griffiths-Schmid attachées à une forme anisotrope du groupe SU(2,2). Ces variétés ont l'avantage, au contraire des variétés de Shimura, de parfois faire apparaître dans leur cohomologie des limites dégénérées de séries discrètes. La première partie étudie ce phénomène dans le cas des limites totalement dégénérées. On prouve que les classes attachées à ces représentations peuvent s'exprimer comme cup-produits d'autres classes attachées à des séries discrètes. La seconde partie étudie les liens entre deux différentes variétés de Griffiths-Schmid obtenues à partir de deux structures complexes. L'une est celle considérée dans la première partie, et l'autre est fibrée holomorphiquement sur une variété de Shimura. On prouve l'existence d'une application bijective entre certains espaces de cohomologie, en s'appuyant sur une interprétation en termes de fonctions holomorphes de la cohomologie de Dolbeault. Ce résultat est généralisé dans l'annexe aux cas des groupes SU(n,n) et SU(n+1,n).

Following his 1924 break with the Paris avant-garde, Francis Picabia (1879-1953) decamped to the French Riviera and soon began work on his radically new Transparency paintings. This series, which occupied Picabia from approximately 1928 to 1933, drew on classical imagery of biblical and mythological subjects, layering disparate human forms and natural motifs in sensuous compositions remarkable for their ambiguous pictorial space and sinuous lines. The Transparencies' resistance to narrative or allegory--notwithstanding their apparent clarity of reference--parallels the paintings' evasion of formal interpretation in spite of their classical beauty; both of these characteristics have made Picabia's Transparencies one of his most inscrutable and misunderstood bodies of work. To avoid treating the Transparencies as a non sequitur or as a conservative abandonment of earlier modernist goals, it is important to understand the sources of the concepts underpinning these works but originating in Picabia's earlier Cubist and Dada periods. Dimensionality, appropriation, figuration, and a rigorous commitment to individualism are all themes from Picabia's acclaimed work in the 1910s and early 1920s that continue into the Transparencies. Particularly relevant are the multivalent interpretations of the spatial fourth dimension--scientific, philosophical, and occult--that Picabia had first encountered in the context of Cubism and the Stieglitz Circle and, later, in his friend Marcel Duchamp's optical experiments. In the Transparencies Picabia's layered outlines both deny linear perspective and suggest projections of interior worlds. In 1936, Picabia affirmed his interest in the fourth dimension, referring specifically to the Transparencies' superimposition at the time he signed Charles Sirató's "Manifeste Dimensioniste." Picabia's visual synthesis of decades of avant-garde concerns in the Transparencies appealed to the American expatriate writer Gertrude Stein, who became one of Picabia's closest friends and confidantes in the early 1930s after she saw his recent paintings. Stein's commentary on Picabia's work and their friendship in "The Autobiography of Alice B. Toklas" and "Everybody's Autobiography" reveals the painter's impact on Stein at a turning point in her career, but also elucidates their shared search for new verbal and visual expressions of the human figure and higher dimensionality. / text

We look at three distinct topics in analysis. In the first we give a direct and easy proof that the usual Newton-Leibniz rule implies the fundamental theorem of algebra that any nonconstant complex polynomial of one complex variable has a complex root. Next, we look at the Riesz representation theorem and show that the Riesz representing measure often can be given in the form of mini sums just like in the case of the usual Lebesgue measure on a cube. Lastly, we look at the idea of holomorphic domination and use it to define a class of complex Banach manifolds that is similar in nature and definition to the class of Stein manifolds.

This study explores the intersection of Modernism and sexology. To date, most studies of sexology’s influence on literature have focused on the importance of inversion in the lesbian salons of interwar Paris and, specifically, on Radclyffe Hall and her associates. The central question in these studies is whether inversion was ultimately beneficial or detrimental to the larger struggle for sexual equality and gay rights. This is an important question and key elements of the debate are reviewed. Sometimes lost in this discussion, however, is sexology’s influence on the creative process of different Modernist writers. By purporting to explain the origins and function of desire, sexology raised the prospect of engineering response, of literally seducing the reader into new aesthetic experiences. These prospects arise not from a literal application of sexological precepts but from a process of critical revision that transformed sexology without undermining the objectivist pretensions upon which the discourse was founded. The dissertation is directed toward explaining the nature of this exchange and its influence on the work of Gertrude Stein, Virginia Woolf, and Djuna Barnes. Theoretically, the study follows Bruno Latour in rethinking the arts/science divide. It suggests writers were able to occupy seemingly self-contradictory positions—embracing both the objective authority of science and the perspectivism of the arts—by exploiting a disavowed hybridity at the heart of the modern condition. This discursive sleight of hand empowered these writers to reinvent both their own identities and the forms in which they worked. Proceeding more or less chronologically, the study begins by looking at Gertrude Stein’s efforts to incorporate the mechanics of attraction into her writing, guided by the work of Otto Weininger. It next examines Virginia Woolf’s exploration of androgyny with reference to Edward Carpenter’s advocacy on behalf of the “intermediate sex”. Finally, attention shifts to Djuna Barnes and the limits of sexology and other attempts to theorize desire. Ultimately, the goal is not to explain sexual difference or to advocate on behalf of any one position. Instead, the dissertation examines how sexology inspired the Modernist imagination in further challenging artistic conventions.

Thesis (Ph. D.)--Rijks-Universiteit te Utrecht, 1923. / Includes bibliographical references (p. [xv]-xx).

This thesis addresses the literary representation of food in the period from 1900 through 1945 in the work of Ford Madox Ford, Gertrude Stein and Virginia Woolf. Taking up nineteenth-century fascinations with sensual and aesthetic taste, these authors explore the implications of food preparation and consumption in Britain, America and France. They use representations of everyday culinary practices as a way to examine articulations of anxiety about the state of civilization, a fear that is amplified and altered by both World Wars. The thesis approaches the question of the significance of food to literary modernism in two ways. The first is a theoretical analysis of modernist ways of thinking about the dialectic between the concepts of civilization and barbarism. The second is grounded in material history, establishing the contexts and conditions of food culture in the first half of the twentieth century. Drawing on sociological thinking from Norbert Elias's conception of the civilizing process and Pierre Bourdieu's theory of distinction, and using a combined methodology of close reading, biographical and historical analysis, I show that food acts as a lens for these authors' ideas about civil society and modernity. My original contribution to knowledge is threefold. The first is my interpretation of 'culinary Impressionism' as an extension and repositioning of current scholarly thinking about Ford's literary Impressionism. The second is my reading of Stein's and Toklas's jointly-authored cookbook draft as evidence of their collaboration. This forms the crux of my argument about Stein adapting domestic culinary techniques into her other writing. The third is in my chapter on Virginia Woolf. My original archival research shows that in A Room of One's Own Woolf's representation of the financial and culinary difference between men's and women's dining in colleges at the University of Cambridge is justified and the material inequality was in fact worse than previously understood. I argue that the disparity in institutional food intensifies Woolf's later reimagining of the term 'civilization' in Three Guineas. While drawing on the work of modernist studies scholars on modernism and the everyday, civilization, and food, my project is unique in demonstrating that food reflects modernist conceptions of civilization and barbarism. My thesis contributes to the understanding of transatlantic aesthetics and gendered productions of modernism by illuminating the centrality of agriculture, cookery, domestic work and institutional dining to modernist authors.