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/ Universidade de Lisb...arrow_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/
addClaim

Estudo sobre a adaptação dos modelos log-lineares à ordinalidade e à presença de zeros amostrais em tabelas de contingência

Authors: Carita, Regina Maria Baltazar Bispo;

Estudo sobre a adaptação dos modelos log-lineares à ordinalidade e à presença de zeros amostrais em tabelas de contingência

Abstract

Os modelos log-lineares são, desde há muito, uma das mais importantes ferramentas na análise de tabelas de contingência, tendo impacto em muitas áreas científicas. Com a proliferação de packages informáticos capazes de executar os cálculos necessários e apresentar os resultados desejados, a utilização destes modelos tem crescido grandemente. Contudo, a análise de bases de dados de dimensões cada vez maiores, tem aumentado o grau de complexidade e de atipicidade da informação recolhida, tornando necessário o desenvolvimento de novas técnicas e a adaptação dos procedimentos clássicos a problemas específicos. Neste trabalho aborda-se a problemática da adaptação dos modelos log-lineares clássicos à ordinalidade das variáveis e à existência de zeros em tabelas de contingência. Neste sentido, são abordados os modelos log-lineares ordinais que permitem descrever padrões de associação e interacção (ou a sua ausência) inerentes à ordinalidade das variáveis. Por outro lado, aborda-se o problema da existência de celas com zeros amostras, na perspectiva de construção de modelos quase-log-lineares. A abordagem das metodologias é estendida ao caso de coexistência de escalas ordinais e presença de zeros, abordando-se os modelos quase-log-lineares ordinais. As metodologias abordadas foram aplicadas a um conjunto de dados enfatizando a sua aplicabilidade na análise de dados reais. Os dados, relativos a escalas nutricionais de crianças e organizados numa tabela tridimensional, permitiram, numa perspectiva de intervenção precoce e preventiva da obesidade infantil, estimar a probabilidade de uma criança ser obesa, ou possuir excesso de peso, num determinado contexto nutricional familiar.

Log-linear models are, since a long time, one of the most used methodologies in contingency tables analysis, with impact n many scientific fields. With the growing availability of statistical packages that perform the necessary calculations and present the so desired results, these models have been increasingly used. Yet the analysis of bigger amounts of data have been incrementing the complexity and the messiness of collected data, implying the development of new technics and the adaptation of classical procedures to specific problems. This study deals with the adaptation of classical log-linear models to ordinal cross classified data with zero counts. These models are adapted to the ordinality of the variables by including association and interaction terms that reflect the hierarchical characteristics of the categorical variables. Quasi-log-linear models are fitted to overcome estimated random zeros. Studied methods were then applied to real data emphasizing the applicability of this methodology in data analysis. Parent and offsprings body structure data organized into a three-way contingency table, allow to determine the probability of a child being obese (or having excess of weight) in a particular nutritional family context.

Tese de mestrado em Probabilidades e Estatística, apresentada à Universidade de Lisboa através da Faculdade de Ciências, 2008

Country
Portugal
Related Organizations
Keywords

Teses de mestrado, Estatística

  • 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