none

No description available.

In 1990 James D. Lewis made a conjecture on the representability of algebraic Chow groups of projective algebraic manifolds. We prove that his conjecture holds for smooth complex complete intersections satisfying a numerical condition and consider some applications to motives. / Mathematics

University, Diss., 1999--Osnabrück.

No description available.

Dans toute cette thèse, k désigne un corps de caractéristique différente de 2 et par variété nous désignons un k-schéma, séparé et de type fini. Nous allons étudier $X(\alpha_1)$ et $X(\alpha_2)$, les variétés homogènes projectives associées à chacune des deux racines d'un groupes de type $G_(2)$. La pemière d'entre elles, $X(\alpha_1)$, est une quadrique projective de dimension 5 associée à une voisine de Pfister et l'autre, $X(\alpha_2)$, est une variété de Fano (de genre 10). Ces deux variétés ne sont pas isomorphes, pourtant elles le deviennent en tant qu'objets d'une catégorie plus large, à savoir la catégorie des correspondances (et par conséquent également dans la catégorie des motifs de Chow). Nous établissons que ce résultat est vrai que les variétés soient déployées ou non. Dans un premier chapitre, nous rappelons quelques résultats classiques sur les algèbres d'octonions et construisons un modèle d'algèbres d'octonions déployée. Dans le second, nous présentons les variétés mises en jeu et rappelons pour cela des notions essentielles de la théorie des groupes algébriques ainsi que de celle des foncteurs de points. Dans le troisième chapitre, nous construisons une structure cellulaire de $X(\alpha_2)$ lorsqu'elle est déployée, étape essentielle de notre travail. C'est également dans ce chapitre que nous calculons les relations définissant la structure d'anneau de $X(\alpha_2)$. Enfin, dans le quatrième et dernier chapitre, nous introduisons la catégorie des correspondances avant de prouver notre théorème de nilpotence dans le cas particulier de la variété $X(\alpha_2)$, puis nous établissons l'isomorphisme motivique en toute généralité.

[pt] O Capital Asset Pricing Model (CAPM) é o modelo mais difundido e utilizado para determinação do custo de capital de empresas e estimação do retorno esperado de ações. Neste modelo, o parâmetro fundamental é o beta, que define a intensidade em que determinado ativo é exposto aos retornos do mercado. O objetivo deste trabalho é avaliar a estabilidade dos betas de uma vasta quantidade de ações no mercado brasileiro em três períodos de alta volatilidade: a crise asiática de 1997, a turbulência no mercado financeiro pré-eleições 2002 e a crise financeira de 2008, nas quais foram analisadas 55, 79 e 172 empresas respectivamente. Cada ciclo de crise foi dividido em três períodos de 52 semanas e os respectivos betas foram comparados utilizando testes de Chow e com regressões com variáveis dummy. Os resultados de ambos os testes foram similares para as crises analisadas, indicando que entre 11 porcento e 27 porcento das empresas apresentaram variação de seus betas, com 5 porcento de significância, quando comparados os períodos pré-crise e durante a crise. Não obstante, ao confrontar períodos pré-crise e pós-crise, a maior parte das empresas que apresentaram variação anteriormente não rejeitaram a hipótese de estabilidade. Estes resultados indicam que, conforme esperado, os betas tendem a ser estáveis no longo prazo. / [en] The Capital Asset Pricing Model (CAPM) is the most widespread model, used to determine the cost of capital of firms and estimate expected stock returns. In this model, the most important parameter is the beta, which defines the magnitude of exposition to market returns, for a particular asset. The objective of this essay is to evaluate beta stability of a vast amount of shares in the Brazilian stock market in three highly volatile periods: the Asian crisis in 1997, the financial market turmoil before 2002 presidential elections and the 2008 financial crisis. The sample included 55, 79 and 172 companies, respectively. Each crisis cycle was then divided into three periods of 52 weeks and then the stability of their betas was measured using regressions with dummy variables and Chow tests. We have reached similar results for the crises analyzed, indicating that between 11 percent and 27 percent of the companies changed their betas when comparing pre-crisis and during crisis periods, with a 95 percent confidence level. Nevertheless, by comparing pre-crisis and post-crisis periods, most of the companies that exhibited change in their betas when entering the crisis, did not reject the stability hypothesis. These results indicate that, as expected, betas tend to be stable in the long term.

nÃo hà / Este estudo teve como objetivo investigar a existÃncia de uma quebra estrutural na relaÃÃo entre o setor bancÃrio e o IBOVESPA durante o perÃodo de 1 de janeiro de 2007 a 29 de julho de 2011, em consequÃncia da crise financeira ocorrida em 2008. Foram empregadas tÃcnicas tradicionais em FinanÃas e Econometria para analisar os impactos da referida crise sobre o setor bancÃrio brasileiro, tomando por base as cotaÃÃes diÃrias de fechamento das aÃÃes dos principais bancos brasileiros: Banco do Brasil, Bradesco, Itaà e do IBOVESPA. Na metodologia utilizou-se o modelo de apreÃamento de ativos, CAPM, na mensuraÃÃo do risco sistÃmico. Observou-se que as evidÃncias estatÃsticas, obtidas com os testes de Chow e teste t para diferenÃa de mÃdias, indicam fundamentalmente, que foi possÃvel captar um efeito diferenciado durante a crise de 2008 entre os bancos privados e o banco pÃblico em relaÃÃo ao risco sistÃmico, alÃm de captar uma mudanÃa estrutural em 24 de outubro de 2008, mudanÃa essa detectada a partir do teste de Chow. / This research aimed to investigate the existence of a structural break in the relationship between the banking sector and IBOVESPA during the period of January 1st 2007 to July 29th 2011, in consequence of the financial crisis occurred in 2008. Traditional techniques were employed in Finance and Econometrics knowledge to analyze the impacts of this crisis on the Brazilian banking sector, based on the daily closing prices of the shares of major Brazilian banks, includes Banco do Brasil, Bradesco, Itaà and IBOVESPA. The methodology used was based on the model of asset pricing, CAPM, in the measurement of systemic risk. It was observed that the statistical evidence, gained with the Chow test and t test for averages differences, basically indicate that it was possible to capture a different effect during the 2008âs crisis between public bank and private banks in relation to systemic risk, and capture a structural change in October 24, 2008, a shift detected from the Chow test.

Soient $X$ et $Y$ des variétés au dessus d’un corps $F$. Dans de nombreuses situations, il s’avère important de savoir si un cycle algébrique modulo équivalence rationnelle y sur Y, défini au dessus du corps des fonctions $F(X)$ de $X$, est en fait déjà défini au niveau du corps de base $F$. Dans cet essai, on traite de cette question, en faisant varier la variété $X$ parmi des variétés telles que des quadriques, des variétés projectives homogènes ou des espaces principaux homogènes. Dans chaque situation, on utilise des outils appropriés tels que les opérations de Steenrod, des résultats de décomposition motivique, ou certains invariants cohomologiques de groupes algébriques. / Let $X$ and $Y$ be some varieties over a field $F$. In many situations, it is important to know if an algebraic cycle modulo rational equivalence $y$ on $Y$ defined over the function field $F(X)$ of $X$ is actually defined over the base field $F$. In this dissertation, we study that matter, making the variety $X$ vary among varieties such as quadrics, projective homogeneous varieties or principal homogeneous spaces. In each situation, we use appropriate tools, such as Steenrod operations, motivic decomposition results or cohomological invariants of algebraic groups.

Abstract Under the great impact and known to all, the structural changes may be obviously clear that can be observed out, when being not so obvious, should observe whether there are structural changes to appear, it is very difficult to get a clear result clearly to turn into. This paper tries to use such models as Chow test, CUSUM test, CUSUMSQ test, STAR in unknown cases, whether going to assay the data separately has structural changes, various kinds of examination ways of result received are each different to some extent, clear conclusion not unanimous and unified. The dominance the Chow test is the most obvious; Consider heteroscedasticity and autocorrelation restriction in because CUSUM test and CUSUMSQ test model, it is not apt to demonstrate the dominance instead; The STAR depends on state of the materials.