Centre for Mathematics, University of Coimbra

Seminars - Historic

A brief introduction to Machine Learning: from the perceptron to large language models

2023-12-15

Speaker: Francisco Pereira (CISUC & Polytechnic Institute of Coimbra)

The field of machine learning has experienced a strong expansion in recent years, mostly driven by the development of new computational methods and the unprecedented surge in the volume of data accessible for processing and analysis. This dynamic has opened up new frontiers for exploration...

A lattice-theoretic approach to submodular mean field games based on stochastic orders

2023-07-19

Speaker: Max Nendel (Univ. of Bielefeld, Germany)

We provide an abstract framework for submodular mean field games and identify verifiable sufficient conditions that allow to prove existence and approximation of strong mean field equilibria in models, where data may not be continuous with respect to the measure parameter and common noise is...

Non symmetric Cauchy kernel, crystals and last passage percolation

2023-06-21

Speaker: Olga Azenhas (CMUC, Univ. Coimbra)

We use non-symmetric Cauchy kernel identities to get the law of last passage percolation (LPP) models in terms of Demazure characters (which are non symmetric polynomials) of a gl_n Lie algebra. The construction is based on restrictions of the Robinson-Schensted-Knuth (RSK)...

Families of polytopes with rational linear precision in higher dimensions

2021-11-03

Speaker: Eliana Duarte (CMUP, Univ. Porto)

Polytopes with rational linear precision are of interest in the Geometric Modeling community because of their approximation properties and it is an open question to classify them in dimension d > 2. This classification question is closely related to discrete statistical models with...

Information in additional observations from a non-parametric experiment

2019-12-18

Speaker: Tilo Wiklund (Univ. of Uppsala, Sweden)

Given some statistical experiment E we may, at least in principle, make multiple, independent, repetitions of it. Say that one statistician repeats the experiment n times, while another repeats it n+1 times. How much better off will the latter be? They certainly should not do be any worse...

A generalized approach to stochastic dominance

2019-11-13

Speaker: Tommaso Lando (Univ. of Bergamo, Italy & VSB-TU Ostrava, Czech Republic)

In the theory of decision under uncertainty, ranking probability distributions with respect to the preferences of decision makers represents a major issue. In this context, the main ranking criteria are the first and the second-order stochastic dominance, FSD and SSD, respectively, due to...

On entropy functionals associated with Bernstein stochastic processes

2019-10-18

Speaker: Pierre Viullermot (Univ. Lorraine & Univ. Lisboa)

Bernstein processes constitute a generalization of Markov processes, and there are many equivalent ways to define them. In this talk I will show how to generate such processes from a hierarchy of forward-backward systems of decoupled deterministic linear parabolic partial differential...

Introduction to causal analysis

2019-05-29

Speaker: Maria Králová (Masaryk Univ., Czech Republic)

Researchers in many areas where only observational data are available often face the challenge of assessing the effect of the treatment on the outcome. Within a framework of traditional statistical analysis of multivariate data, only different measures of association are available....

Information lost when resampling

2018-12-12

Speaker: Tilo Wiklund (Univ. of Uppsala, Sweden)

Various procedures in statistics rely on taking decisions based on data that has been resampled (usually with replacement). Examples include bootstrap-based confidence bounds and error estimates or "bagged" estimators. Since resampling is an additional source of randomness such...

Nonparametric first-order analysis of spatial and spatio-temporal point processes. Application to wildfire patterns

2018-01-19

Speaker: Wenceslao González Manteiga (Univ. Santiago de Compostela, Spain)

Spatial point patterns arise in a wide variety of scientific contexts, including seismology, forestry, geography and epidemiology. Wildfire is the most ubiquitous natural disturbance in the world and represents a problem of considerable social and environmental importance; particularly, in...

Some results on stochastic ageing and ordering

2017-12-15

Speaker: Idir Arab (CMUC, Univ. Coimbra)

The concepts of stochastic ageing and ordering, which play an important role in any reliability analysis, describe how a component or system improves or deteriorates with age, and how to decide which component or system is ageing faster (in some sense). Stochastic ordering of distributions...

Some results on dependent random variables

2017-11-24

Speaker: Milto Hadjikyriakou (Univ. of Central Lancashire, Cyprus)

In recent years, concepts of dependence have been the focus of substantial research activity and applications of dependent random variables can be found in many different fields. In the first part of this talk some classes of dependent random objects are presented together with various...

Early exercise boundaries for American-style knock-out options

2017-05-18

Speaker: José Carlos Dias (ISCTE-IUL Business School)

Under mild assumptions, this paper proposes a novel representation for the early exercise boundary of American-style double knock-out options in terms of the simpler optimal stopping boundary of a nested single barrier American-style contract. Therefore, and from now on, any single-factor...

Characterization and estimation of the upcrossings index

2017-05-17

Speaker:

Ana Paula Martins (CMA-UBI, Univ. da Beira Interior)

Extreme Value Theory aims to predict occurrence of rare events, which in numerous situations have disastrous impacts, making an adequate estimation of the parameters related with such events of primordial importance in Statistics of Extremes. The extremal index and the upcrossings index play...

Nonparametric models for extreme risks with urn models

2017-03-22

Speaker: Juerg Husler (Univ. of Bern, Switzerland)

Extreme risk is often modelled with parametric or semi-parametric models. Here we present nonparametric models for a large setup of risks where the underlying process is not stationary, but is changing with time. The approach is based on the flexible class of urn models. This is made...

Statistical detection of some topological and geometric features

2017-01-26

Speaker: Antonio Cuevas (Univ. Autónoma de Madrid, Spain)

The subject of this talk is closely related to the theories of set estimation and manifold estimation. Our object of interest is a, possibly lower-dimensional, compact set M in Rd. The general aim is to identify (via stochastic procedures) some qualitative or quantitative features of M of...

On ARL-unbiased c-charts for INAR(1) Poisson counts

2017-01-13

Speaker: Manuel Cabral Morais (CEMAT, Univ. Lisboa)

Counts of nonconformities are frequently assumed to have a Poisson distribution. The integer and asymmetrical character of this distribution and the value of its target mean may prevent the quality control operator to deal with a chart with a pre-specified in-control average run length (ARL)...

Stochastic dynamics on point processes

2016-09-30

Speaker: Antonio Sodre (Department of Mathematics, Univ. of Texas at Austin)

Deterministic dynamics on stationary point process in R^d are built upon compatible point-shifts: translation invariant mappings from each point of the process to another. When a point-shift is applied multiple times to a point-process it creates a sequence of distributions, namely,...

Application of the Kalman filter in the statistical modelling of time series

2016-06-17

Speaker: Marco Costa (ESTGA & CIDMA, Univ. Aveiro)

The Kalman Filter (KF), initially submitted by the Hungarian Rudolf E. Kalman in the context of control problems, has been widely applied in statistical modelling of stochastic processes that allow a state space representation. The success of the KF is attributed to the optimal estimators...

Mixed-Effect State Space models applied to environmental data: the discrimination of the Water Quality Monitoring Sites in River Vouga

2016-04-15

Speaker: Magda Monteiro (ESTGA & CIDMA, Univ. Aveiro)

The surface water quality monitoring is an important concern of public organizations due to its relevance for public health. Statistical methods are taken as consistent and essential tools in the monitoring procedures in order to prevent and identify environmental problems. This work...

The Taylor property in non-negative autoregressive stochastic processes

2015-11-13

Speaker: Cristina Martins (DMUC, Univ. Coimbra)

In this talk we analyse the presence of the Taylor property in linear models. Limiting the study to non-negative models, we begin by recalling the main theoretical results on the occurrence of the Taylor property in autoregressive models of order one. These results are discussed in models...

A branching process approach to the survival of populations: skeleton process and stochastic introgression

2015-07-17

Speaker: Maria Conceição Serra (CMAT, Univ. Minho)

In this talk we present branching processes as a tool to model the evolution of populations which are, in principle, doomed to extinction but manage to survive. Survival is usually achieved through the appearance of different types of individuals that give an “advantage” to the...

Convergence of Kiefer-Wolfowitz algorithm under quasi-associated random errors

2015-06-18

Speaker: Idir Arab (University of Bejaia, Algeria)

In this talk, we study the algorithm of Kiefer-Wolfowitz under quasi-associated random errors. We establish the complete convergence and obtain an exponential bound. Additionally, we build a confidence interval for the minimum. Numerical examples are sketched out to confirm the theoretical...

Modelling with clinical data: examples, limitations, aims ...

2015-03-11

Speaker: Carla Henriques (CMUC and Instituto Politécnico de Viseu)

In this seminar, we will present some examples of applications of Statistics in the field of Medicine. Research questions are posed by health professionals, who aim to see their "sensibilities" supported statistically. The answers are not always straightforward, and the...

Modeling and analysis of count data time series: recent research activities

2015-02-03

Speaker: Christian H. Weiß (Helmut Schmidt Univ. Hamburg, Germany)

After a brief introduction to basic concepts and examples of count data time series, we shall focus on two particular models and their statistical analysis: the (compound) Poisson INAR(1) model and the (self-exciting threshold) binomial AR(1) model.The Poisson INAR(1) model is the most...

Density-based cluster analysis: an interface between Statistics, Differential Topology and Dynamical Systems

2014-12-10

Speaker: José E. Chacón (Univ. de Extremadura, Spain)

Despite its popularity, the investigation of some theoretical aspects of clustering has been relatively sparse. One of the main reasons for this lack of theoretical results is surely the fact that, unlike the situation with other statistical problems as regression or classification, for some...

Random fields and random sampling

2014-11-14

Speaker: Maria da Graça Temido (CMUC, Univ. Coimbra)

This talk is focused on the study of the maximum term of a stationary bivariate random field. Considering the double random dimensions with asymptotic geometric growing pattern, in probability, a mixture distribution for the maximum is established. ...

Extreme Value Theory for chaotic dynamics

2014-10-31

Speaker: Jorge Milhazes Freitas (CMUP, Univ. Porto)

We will give an introduction to the Extreme Value Theory in a dynamical systems context. In particular, we will interpret the occurrence of rare events in terms of the hitting time to certain shrinking targets on the phase space of the possible states of a given system. We will present new...

A new class of integer valued processes with conditional heteroscedasticity

2014-09-19

Speaker: Filipa Silva (CMUC)

The aim of this talk is to introduce an integer-valued model with conditional distribution belonging to the class of compound Poisson probability laws. This proposal includes, in particular, the Poisson INGARCH model, and the negative binomial and generalized Poisson models. We show the...

Moderate deviations for dependent variables

2014-06-06

Speaker: Paulo Eduardo Oliveira (CMUC)

Large and moderate deviations are an important topic in many theoretical or applied statistical areas. We start by a brief overview of large deviations, mentioning the results for dependent variables. A special interest is put for exact convergence rates. After a first quick look at exact...

Taylor property revisited

2014-03-28

Speaker: Joana Leite (CMUC and Instituto Politécnico de Coimbra, ISCAC)

Empirical properties systematically observed in time series of a certain field play an important part in scrutinizing models, as models proposed for them should be able to reproduce these, so called, stylized facts. The Taylor effect is a stylized fact regarding the autocorrelation...

On closure properties of some risk aversion measures

2013-12-17

Speaker: Nuria Torrado-Robles (CMUC, Univ. Coimbra)

Increasing risk aversion (IRRA) is a common assumption in economics, but this property often is difficult to verify. In this talk, we provide alternative characterizations of IRRA utility functions in terms of increasing absolute risk aversion (IARA) utility functions. We also present some...

Financial risk measures

2013-11-26

Speaker: Miguel Mendes (FEUP/CMUP, Univ. Porto)

We review the concept of risk measure and its use as a capital requirement in the financial industry or as a premium in the insurance context. We then consider the Choquet integral as an important representation of risk measures and specialise to distortion risk measures. Our main innovation...

Modelação e monitorização de dados usando famílias de distribuições

2013-10-04

Speaker: Fernanda Otília Figueiredo (Fac. Economia, Univ. Porto)

A utilização de famílias de distribuições na modelação de dados reais permite uma maior versatilidade na escolha do modelo que melhor se ajusta às diferentes peculiaridades dos dados. Apesar da desvantagem subjacente à...

A new test for multivariate normality by combining extreme and non-extreme BHEP test statistics

2013-10-04

Speaker: Carlos Tenreiro (CMUC/FCTUC)

In this talk we present a new multiple test procedure for assessing multivariate normality which combines the BHEP (Baringhaus-Henze-Epps-Pulley) tests by considering extreme and non-extreme choices of the smoothing parameter figuring in the definition of the BHEP test statistic. The results...

The minimum description length in line transect sampling

2012-01-20

Speaker:

Russell Alpizar-Jara (Univ. Évora)

Population density estimation in line transect sampling requires fitting a probability density function denoted by f(y|Ѳ), where y represents the perpendicular distance from a detected animal (or object) to a transect line, and Ѳ represents the vector parameter indexing this...

Integer-valued bilinear time series models

2011-11-29

Speaker: Isabel Pereira (Univ. Aveiro)

The analysis of count processes has become an important area of research in the last two decades. In this paper we consider the simple non-negative integer-valued bilinear process INBL(1,0,1,1). The parameter estimation is addressed and the problem of forecasting integer-valued time...

Moment analysis of distributions: classical and recent results

2011-09-20

Speaker: Jordan Stoyanov (Newcastle Univ., UK)

The main discussion will be on distributions and their characterization in terms of the moments. There are distributions which are unique (M-determinate) and others which are non-unique (M-indeterminate). When analyzing specific stochastic model we clarify why in some cases a distribution...

Estimating the adjustment coefficient in Risk Theory

2011-09-16

Speaker: Ana Cristina Moreira de Freitas (Fac. Economia, Univ. Porto)

We consider the problem of estimating the adjustment coefficient R in the Sparre Andersen model, which allows us to estimate upper bounds for the ruin probability of insurance companies.We propose a consistent estimator for R and establish a result about its asymptotic normality. Moreover,...

Exploration of genomewide data: future challenges and the state of the art

2010-10-06

Speaker: Magnus Fontes (Lund University, Sweden)

I will present a collection of statistical and mathematical methods useful for the exploration of some types of multivariate data and in practice show how we have implemented them in the commercial software Qlucore Omics Explorer (QOE). In particular I will discuss classical principal...

High and massive excursions of Gaussian processes

2010-10-04

Speaker:

Juerg Husler (Univ. Bern, Switzerland)

Later...

Estimação adaptativa e invariante do índice de valores extremos

2010-05-17

Speaker: Maria Ivette Gomes (Univ. Lisboa)

Iremos abordar um método de estimação semi-paramétrica de viés-reduzido de um índice de valores extremos (EVI, do Inglês, extreme value index) positivo. Neste contexto de caudas-direitas pesadas, os estimadores clássicos do EVI...

Trees and Poisson Point Processes

2008-09-12

Speaker: Iesus Diniz (USP, Brasil)

A a connected graph and without cycles, a tree is constructed, with a single infinite self-avoiding path, an end, and the vertices of the tree are points of an infinite sequence of Independent Poisson Point Processes defined in R^d, such that for every k larger or equal than 1, the rate...

A propriedade de Taylor nos modelos TARCH

2008-07-18

Speaker: Joana Leite (Instituto de Contabilidade e Administração de Coimbra, IPC)

Na análise de séries temporais, a selecção do modelo que melhor se ajusta aos dados é determinante para validar a inferência estatística e a previsão. Esta escolha é baseada, em particular, no nosso conhecimento das propriedades teóricas do modelo, uma vez que escolhemos aquele que melhor...

Empirical portfolio selection

2008-06-25

Speaker: László Györfi

Area(s): ...

Na natureza nada se perde, tudo se transforma. E em Estatística?

2008-04-11

Speaker: Andreia Hall (Univ. Aveiro)

Neste seminário iremos analisar de que forma a perda de informação se reflecte nas propriedades extremais de uma sucessão estacionária de variáveis aleatórias. Será que a perda de informação (querendo dizer com isso que...

Leis max-semiestáveis como limite do máximo em modelos MA e max-AR de variáveis inteiras

2008-02-07

Speaker: Maria da Graça Temido (CMUC & FCTUC)

Analisamos a convergência em distribuição do máximo, linearmente normalizado, em modelos estacionários de médias móveis e max-autorregressivos de variáveis inteiras. Consideramos o caso particular em que a função de distribuição marginal não pertence a nenhum domínio de atracção max-estável,...

A breakpoint analysis for clustering

2007-12-05

Speaker: Maria Teresa Gallegos (University of Passau, Germany)

Area(s): ...

Graph and Information-theoretic Methods for Analyzing Complex Networks

2007-11-29

Speaker: Mathias Dehmer (Vienna University of Technology)

The analysis of graph-based systems is currently an ongoing and outstanding research topic. It turned out that especially techniques for measuring graph similarity are extremely important. In this talk I will outline some existing and novel methods to measure the structural similarity of...

Limite hidrodinâmico de alguns sistemas de partículas

2007-07-09

Speaker: Patrícia Gonçalves, USP, Brasil

Area(s): ...

Calibração não paramétrica na estimação da idade

2005-07-12

Speaker: Rui Martins (Colégio S. Martinho, Coimbra)

Area(s): ...

Amostragem por distâncias ??? conceitos gerais e uma alternativa na presença de gradientes de densidade

2005-06-21

Speaker: Tiago Marques (CEAUL)

A amostragem por distâncias é um dos métodos mais amplamente utilizados na estimação da abundância de populações animais, sendo usada em quase todos os grupos animais. As distâncias dos transectos aos animais detectados são modeladas, permitindo obter estimativas não enviesadas da abundância...

Regressão linear múltipla com alguns erros correlacionados: métodos clássicos e robustos

2005-05-24

Speaker: Ana Pires (CEMAT, IST)

Este trabalho teve como motivação principal a análise de um conjunto de dados de medicina. Esses dados resultaram de um estudo observacional destinado a identificar os factores que influenciam o resultado de um determinado método de tratamento cirúrgico da escoliose (curvatura lateral...

Detecção, no genoma humano, da influência de dados loci na susceptibilidade para doenças complexas

2005-04-26

Speaker: Teresa Ferreira (CMUC, DMUC)

Area(s): ...

Estimação multietápica de certas funcionais da densidade de probabilidade e suas derivadas

2003-07-02

Speaker: Carlos Tenreiro

Area(s): ...

Uma pequena viagem em análise discriminante discreta: dos modelos de referência à combinaçâo de modelos

2003-05-14

Speaker: Ana Sousa Ferreira (FPCE, Universidade de Lisboa)

Os métodos de análise discriminante aplicam-se a populações munidas duma partição definida a priori, descritas por diversas variáveis explicativas (preditoras). O principal objectivo destes métodos é discriminar/diferenciar as classes da partição, através das características definidas pelas...

Loi limite de l'erreur L1 d'un estimateur de frontiere

2003-03-05

Speaker: Pierre Jacob (Univ. Montpellier II, France)

Area(s): ...

Statistics and physical anthropology : application of the bayesian approach to improve age at death assessment on the skeleton

2003-02-26

Speaker: Aurore Schmitt (Centre Anthropologie, Univ. Bordeaux I, France)

One of the areas where physical anthropologists need to apply more statistical tools is in the estimation of age at death on the skeleton remains. Since the age estimated is the biological age which is different from the chronological one, each result is always associated with a standard...

Novos desenvolvimentos em análise de ruído branco Poissoniano

2003-01-15

Speaker: Maria João Oliveira (GFMUL/Univ. Aberta)

Area(s): ...

Uma nova classe de estimadores do índice de cauda

2002-11-13

Speaker: Frederico Caeiro (Dep. Matemática e C.M.A., F.C.T., Universidade Nova de Lisboa)

Muitos dos estimadores semi-paramétricos do índice de variação regular gama0 dependem da escolha do número k de estatísticas ordinais superiores. Para muitos dos estimadores existentes na literatura (Hill, Momentos e Pickands) a escolha do nível ???óptimo???(nível que minimiza o erro...

Lois limites d'estimateurs fonctionnels de frontiÚres

2001-11-27

Speaker: Pierre Jacob (Université de Montpellier II, França)

Area(s): ...

Estimation de directions revelatrices: quelques résultats récents

2001-06-21

Speaker: M. Delecroix (ENSAE, Paris, France)

Area(s): ...

Comportement asymptotique de processus gaussiens stationnaires d'aprÚs Olivier Garat

2000-11-29

Speaker: Raymond Moché (Université des Sciences et Techniques de Lille, France)

Area(s): ...

La méthode CART: ''Classification and regression tree''

2000-11-14

Speaker: Claude Langrand (Université des Sciences et Techniques de Lille, France)

Area(s): ...

Inférence asymptotique dans les modÚles à mémoire longue

2000-11-06

Speaker: Marc Hallin (Université Libre de Bruxelles, Belgium)

Area(s): ...

Processus autorégressifs Banachiques. Théorie et applications

2000-10-31

Speaker: Denis Bosq (Université Paris VI, France)

Area(s): ...

Introdução às redes de Jackson

2000-09-27

Speaker: Teresa Ferreira (Dep. Matemática, Univ. Coimbra)

Definição de Rede de Jackson. Redes de Jackson abertas e fechadas. Dedução das equações de equilíbrio global e sua solução. Area(s): ...

Leis max-semiestáveis e log-semicÎncavas

2000-07-20

Speaker: M. Graça Temido (Departamento de Matemática, Universidade de Coimbra, Portugal)

A classe das funções de distribuição max-semiestáveis -classe MSS- surgiu recentemente na Teoria de Valores Extremos, associada ao limite em distribuição do máximo, linearmente normalizado, de $k_n$ variáveis aleatórias independentes e identicamente distribuídas, onde $\{k_n\}$ é uma...

Semiparametric Efficiency, Distribution-Freeness, and Invariance

2000-07-05

Speaker: Bas Werker (Tilburg University, Netherlands)

Semiparametric models typically involve a finite-dimensional parameter of interest $\thetab\in\Thetab\subseteq\R^K$, along with an infinite-dimensional nuisance parameter~$f$. Quite often, the submodels corresponding to a fixed value of $\thetab$ possess a group structure that induces a...

Sobre o erro quadrático integrado do estimador automático do núcleo duma densidade de probabilidade

2000-05-03

Speaker: Carlos Tenreiro (Departamento de Matemática, Universidade de Coimbra, Portugal)

A janela hn do estimador do núcleo duma densidade de probabilidade, é normalmente escolhida com base nas observações realizadas. Para um tal estimador, dito automático, descrevemos o comportamento assintótico do seu erro quadrático integrado. Area(s): ...

Uma desigualdade exponencial para variáveis associadas

2000-03-22

Speaker: Carla Henriques (Escola Superior de Tecnologia de Viseu and CMUC, Portugal)

Aborda-se a questão da caracterização da velocidade de convergência em teoremas limites centrais para variáveis aleatórias associadas. Os resultados em estudo baseiam-se nas técnicas de aproximação por variáveis independentes, sendo esta aproximação controlada à custa de desigualdades sobre...

Uma desigualdade de Berry-Esséen para variáveis associadas

2000-03-15

Speaker: Paulo Oliveira (Departamento de Matemática, Universidade de Coimbra, Portugal)

Análise assintótica nos "Threshold models"

2000-01-05

Speaker: Ana Paula Martins Nascimento (Departamento de Matemática, Universidade de Coimbra, Portugal and Université Libre de Bruxelles, Belgium)

Nesta apresentação estimamos de forma eficiente o parâmetro autorregressivo no "Semiparametric Threshold Ar(1) model". Area(s): ...

Quem é que?: um desafio à Estatística - Problemas de autoria em "Novas Cartas Portuguesas"

1999-07-07

Speaker: Maria Madalena de Freitas Malva (Escola Superior de Tecnologia de Viseu, Portugal)

Area(s): ...

Função de autocorrelação estendida generalizada: contributo para a identificação dos modelos de função tranferência

1999-06-29

Speaker: Cristina Oliveira (ISCAP, Portugal) and Daniel MÌller (Universidade Técnica de Lisboa, Portugal)

Nas aplicações, a identificação de um modelo de função transferência é tradicionalmente realizada através da análise da função de correlação cruzada entre as séries input e output. A prática tem mostrado que este instrumento de identificação não é , em geral, suficiente, apresentando um...

Modelos espaciais para análise de proximidades

1999-06-23

Speaker: João Branco (Instituto Superior Técnico, Portugal)

A obtenção de medidas de semelhança e dissemelhança entre elementos de um conjunto de indivíduos, objectos ou estímulos é uma prática corrente em psicologia, sociologia, estudos de mercado e em outras disciplinas. A análise deste tipo de dados tem como objectivos principais a construção de...

Simulação de séries temporais a partir de alguns modelos de espectros

1999-05-25

Speaker: Luísa Canto e Castro (DEIO-Universidade Lisboa, Portugal)

Area(s): ...

Detecção de clusters de doenças raras - Uma incursão no mundo bayesiano

1999-04-12

Speaker: Maria Lucília Carvalho (DEIO-Universidade Lisboa, Portugal)

Apresentam-se, de forma muito geral, os principais métodos estatísticos para a detecção de clusters de doenças raras. Consideram-se clusters temporais, espaciais e espaço-temporais. Ilustra-se a aplicação de dois métodos espaço-temporais: o método de Raubertas e o método de Symons, Grimson &...