Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Estudo Geralarrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Estudo Geral
Master thesis . 2022
Data sources: Estudo Geral
addClaim

Estatística de eventos extremos - Até onde poderemos saltar?

Authors: Marques, Ricardo André Oliveira;

Estatística de eventos extremos - Até onde poderemos saltar?

Abstract

Na origem da teoria estatística de extremos está o Teorema do limite extremal ou Teorema de Gnedenko. Este resultado fundamental estabelece que, em condições bastante gerais, a classe dos possíveis limites em distribuição do máximo de n variáveis aleatórias, reais independentes e identicamente distribuídas, coincide com a classe das distribuições de valor extremo (ou max-estáveis). Esta classe divide-se nas três sub-classes das funções de distribuição de Fréchet, Weibull e Gumbel.A motivação para este trabalho foi o conteúdo do artigo "How far can man go?" ([6]). A partir de uma coleção de dados com as maiores marcas de saltos em comprimento (masculino), os autores justificam o facto de o record (8.95m) obtido pelo atleta Mike Powel em 1991 ainda se manter, uma vez que com estes dados estimam um salto máximo de aproximadamente 9m.Este trabalho contém uma introdução à teoria estatística de valores extremos, onde o Teorema do Limite Extremal é motivado e demonstrado. A caraterização dos domínios de atração das três distribuições de valor extremo é baseada e complementada com a introdução de condições adequadas sobre a função quantil. Para o índice de cauda, parâmetro que permite decidir qual das três distribuições de valor extremo está subjacente aos dados, são apresentados vários estimadores e provada a sua consistência. Ao papel fundamental do índice de cauda juntam-se os testes estatísticos de escolha de domínios de atração. Esta análise permite estimar quantis elevados e o extremo superior do suporte da variável em estudo, quando este é finito.O trabalho termina com duas aplicações a dados extremos, nomeadamente, aos dados do triplo salto feminino e aos tempos mínimos de resolução do cubo de Rubik.

At the origin of the statistical theory of extremes is the Gnedenko's theorem. This fundamental result establishes that, under very general conditions, the class of possible limits in distribution of the maximum of n random variables, independent and identically distributed, coincides with the class of max-stable distributions (or extreme value).This class is divided into the three sub-classes of the Frechét, Weibull and Gumbel distribution functions.The motivation for this work was the paper "How far can man go?" ([6]). From a colection of 720 top observations of long jump male, the authors justify the fact that the record of Mike Powel (8.95m) is still maintained, since with this data they estimate a maximum jump of approximately 9m.This work contains an introduction to the statistical theory of extreme values, where the Gnedenko's Theorem is demonstrated. The characterization of the three domains of atraction is based and complemented with the introduction of appropriate conditions on the quantile function. For the tail index, a parameter that decides of which of the three extreme value distributions is underlying the data, several estimators are presented and their consistency is proven. The fundamental role of the tail index is added to the statistical tests of choice of domains of attraction. This analysis allows estimating high quantiles and the right endpoint of the support of the variable under study, when it is finite. The work ends with two applications to extremal data, namely the data of the female triple jump (outdoor) and the minimum times of realization of the Rubik's cube.

Dissertação de Mestrado em Matemática apresentada à Faculdade de Ciências e Tecnologia

Country
Portugal
Related Organizations
Keywords

Statistics of extremes, Estatística de extremos, Inferência estatística, Limit Theorems, Teoremas limite, Statistical inference

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green