Modelos log-lineares em tabelas de contingência

Os modelos log-lineares permitem enriquecer bastante a análise e interpretação das tabelas de contingência. Embora a nível teórico a sua importância já tenha sido reconhecida há bastante tempo, a nível da sua aplicação prática só há relativamente pouco tempo é que foi reconhecida, devido, sobretudo, às dificuldades de cálculo que lhe são inerentes e que só se resolveram completamente com o desenvolvimento dos computadores e do software adequado. Neste trabalho apresentam-se os métodos básicos da análise log-linear de tabelas de contingência bidimensionais e tridimensionais.

Modelling and Forecasting Mortality Patterns

Thesis submitted in partial fulfillment of the requirements for the Degree of Doctor of Statistics and Information Management

Continuidade de cuidados numa Unidade de Saúde Familiar : o efeito nos custos

RESUMO - Caracterização do problema: O sistema de saúde português atingiu um patamar de ineficiência tal que urge ser reestruturado de forma a torná-lo sustentável. De forma a atingir este nível de sustentabilidade, uma série de soluções podem ser consideradas das quais destacamos a integração de cuidados. Este conceito exige que os diferentes níveis de saúde sigam um único caminho, trabalhando de forma coordenada e contínua. A integração de cuidados pode ser implementada através de várias tipologias entre as quais se destaca a integração clínica que por sua vez é composta pela continuidade de cuidados. Assim, ao medir a continuidade de cuidados, quantifica-se de certa forma a integração de cuidados. Objetivos: Avaliar o impacto da continuidade de cuidados nos custos. Metodologia: Os dados foram analisados através de estatísticas descritivas para verificar o seu grau de normalidade. Posteriormente foram aplicados testes t-student para analisar a existência de diferenças estatisticamente significativas entre as médias das diferentes variáveis. Foi então estudado o grau de associação entre variáveis através da correlação de spearman. Por fim, foi utilizado o modelo de regressão log-linear para verificar a existência de uma relação entre as várias naturezas de custos e os índices de continuidade. Com base neste modelo foram simulados dois cenários para estimar o impacto da maximização da continuidade de cuidados nas várias naturezas de custos. Conclusões: No geral, verifica-se uma relação muito ligeira entre a continuidade de cuidados e os custos. Mais especificamente, uma relação mais duradoura entre o médico e o doente resulta numa poupança de custos, independentemente da tipologia. Analisando a densidade da relação, observa-se uma relação positiva entre a mesma e os custos totais e o custo com Meios Complementares de Diagnóstico e Terapêutica (MCDT). Contudo verifica-se uma relação médico-doente negativa entre a densidade e os custos com medicamentos e com pessoal. Ao analisar o impacto da continuidade de cuidados nos custos, conclui-se que apenas a duração da relação médico-doente tem um impacto negativo em todas as categorias de custos, exceto o custo com medicamentos. A densidade de cuidados tem um impacto negativo apenas no custo com pessoal, influenciando positivamente as outras categorias de custos. Extrapolando para o nível nacional se o nível de densidade de uma relação fosse maximizado, existiria uma poupança de 0,18 euros, por ano, em custos com pessoal.

Extremum estimators and stochastic optimization methods

Submitted in partial fulfillment for the Requirements for the Degree of PhD in Mathematics, in the Speciality of Statistics in the Faculdade de Ciências e Tecnologia

Interfaces for television content sharing and annotation

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Past/present and future worries: na fMRI study

For the past decade, numerous imaging techniques gave rise to remarka-ble progresses in the understanding of brain’s structure and function. Amongst the wide variety of studies onto the field of neuroscience, neuropsychiatric re-searches with resource to neuroimaging have attracted increasing attention. The present study will focus on the identification of brain areas recruited while normative subjects read sentences related to past/present or future wor-ries. Our main aim was to accurately characterize these brain areas while providing them with a time-stamp that would hopefully help us understand the implications of past/present memories and future envisioning in worrying episodes. With that purpose, functional magnetic resonance imaging data was collected from ten healthy individuals. The obtained data was processed and statistically treated using the General Linear Model and both Fixed and Ran-dom Effects Analysis for group-level results. Thereafter, a Multi-Voxel Pattern Analysis with Searchlight Mapping was performed in order to find patterns of activation that allow differentiation between conditions. The obtained results indicate higher brain activation while reading sen-tences related to past/present worries when compared to future worry or neu-tral sentences. The main areas include frontal cortex, posterior parietal, occipital and temporal areas. Worrying, per se, was characterized by activation of the medial posterior parietal cortex, left posterior occipital lobe and left central temporal lobe. With the searchlight mapping approach we were able to further identify patterns of distinction between conditions, which were located in the parietal, limbic and frontal lobes.

Location analysis of city sections: socio-demographic segmentation and restaurant potentiality estimation : a case study of Lisbon

One of the objectives of this study is to perform classification of socio-demographic components for the level of city section in City of Lisbon. In order to accomplish suitable platform for the restaurant potentiality map, the socio-demographic components were selected to produce a map of spatial clusters in accordance to restaurant suitability. Consequently, the second objective is to obtain potentiality map in terms of underestimation and overestimation in number of restaurants. To the best of our knowledge there has not been found identical methodology for the estimation of restaurant potentiality. The results were achieved with combination of SOM (Self-Organized Map) which provides a segmentation map and GAM (Generalized Additive Model) with spatial component for restaurant potentiality. Final results indicate that the highest influence in restaurant potentiality is given to tourist sites, spatial autocorrelation in terms of neighboring restaurants (spatial component), and tax value, where lower importance is given to household with 1 or 2 members and employed population, respectively. In addition, an important conclusion is that the most attractive market sites have shown no change or moderate underestimation in terms of restaurants potentiality.

Estimation and hypothesis testing in mixed linear models

Dissertação apresentada para obtenção do Grau de Doutor em Matemática, Estatística, pela Universidade Nova de Lisboa, faculdade de Ciências e Tecnologia

A fractional linear system view of the fractional brownian motion

Nonlinear Dynamics, Vol. 38

Prognostic modelling of breast cancer patients: a benchmark of predictive models with external validation

Dissertação apresentada para obtenção do Grau de Doutor em Engenharia Electrotécnica e de Computadores – Sistemas Digitais e Percepcionais pela Universidade Nova de Lisboa, Faculdade de Ciências e Tecnologia

Development and validation of a risk model for prediction of hazardous alcohol consumption in general practice attendees : the PredictAL study

Background: Little is known about the risk of progression to hazardous alcohol use in people currently drinking at safe limits. We aimed to develop a prediction model (predictAL) for the development of hazardous drinking in safe drinkers. Methods: A prospective cohort study of adult general practice attendees in six European countries and Chile followed up over 6 months. We recruited 10,045 attendees between April 2003 to February 2005. 6193 European and 2462 Chilean attendees recorded AUDIT scores below 8 in men and 5 in women at recruitment and were used in modelling risk. 38 risk factors were measured to construct a risk model for the development of hazardous drinking using stepwise logistic regression. The model was corrected for over fitting and tested in an external population. The main outcome was hazardous drinking defined by an AUDIT score >= 8 in men and >= 5 in women. Results: 69.0% of attendees were recruited, of whom 89.5% participated again after six months. The risk factors in the final predictAL model were sex, age, country, baseline AUDIT score, panic syndrome and lifetime alcohol problem. The predictAL model's average c-index across all six European countries was 0.839 (95% CI 0.805, 0.873). The Hedge's g effect size for the difference in log odds of predicted probability between safe drinkers in Europe who subsequently developed hazardous alcohol use and those who did not was 1.38 (95% CI 1.25, 1.51). External validation of the algorithm in Chilean safe drinkers resulted in a c-index of 0.781 (95% CI 0.717, 0.846) and Hedge's g of 0.68 (95% CI 0.57, 0.78). Conclusions: The predictAL risk model for development of hazardous consumption in safe drinkers compares favourably with risk algorithms for disorders in other medical settings and can be a useful first step in prevention of alcohol misuse.

Stratigraphic interpretation of Well-Log data of the Athabasca Oil Sands of Alberta Canada through Pattern recognition and Artificial Intelligence

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.