Recent seminars

Europe/Lisbon
Online

Milan Stehlík, Institute of Statistics, Universidad de Valparaíso, Chile

From data transformations and aggregations to transfer functions: The importance of extremes and boundary data models

Biological, physical, and ecological systems offer a lot of complexity that should be well understood before valuable interventions can be made. We will address both complex and extreme measurements from these systems. There is a necessity to classify appropriate learning mechanisms and define transfer functions and statistics. A natural question may arise: how to address extreme parts of data? How to define boundaries of the datasets and what can be the effects on statistical properties of estimated structures (e.g. uniqueness of copulas? Can we provide efficient estimators of extremes? For closed physical systems, all can be well integrated into both natural and technical sciences, which gives us an optimal instrument for the decomposition of data into stochastic, deterministic, and chaotic part. In particular, we will introduce SPOCU transfer function and provide some of its unique properties for processing of complex data, statistical learning will be discussed, and tuning of parameters of SPOCU-based neural networks will be explained. During the talk, I will acknowledge the contributions of the Portuguese Extreme group and outline some relations to t-Hill-based estimators. The t-Hill approach will be introduced from a robustness perspective, mentioning and interconnecting with articles, among others. Attractive applications to biological systems, such as mass balance in the ecosystem of glaciers in Patagonia, or methane emissions from wetlands will be addressed.

Joint seminar CEMAT and CEAUL

Europe/Lisbon
Online

Joaquin Cavieres, University of Gӧttingen, Germany

Approximated Gaussian random field under different parameterizations for MCMC

Fitting spatial models with a Gaussian random field as spatial random effect poses computational challenges for Markov Chain Monte Carlo (MCMC) methods, primarily due to two factors: computational speed and convergence of chains for the hyperparameters. To deal with this, a Gaussian random field can be approximated by a Gaussian Markov random field using stochastic partial differential equations. This methodology is commonly used in “latent Gaussian models”, where the inference is done by the Integrated Nested Laplace Approximations, but rarely used in an MCMC method. In this contribution, we evaluated different parameterizations of the approximated Gaussian random field, specifically using the Hamiltonian Monte Carlo algorithm of the Stan software. A simulation study demonstrated that models using the hyperparameters ρ and σu were better able to estimate the values used to simulate the spatial random field. Their speed computation were faster compared to models parameterized with κ and τ. In real data application, the index of relative abundance estimated for Pollock indicates similar trends for the six models proposed. However, models incorporating ρ and σu demonstrated faster computation compared to those utilizing κ and τ, corroborating the results found in the simulation. Even more important, none of these models encountered convergence issues, as indicated by the Rhat statistic.

Joint seminar CEMAT and CEAUL

Europe/Lisbon
SASlab (6.4.29) Faculty of Sciences of the Universidade de Lisboa — Online

Miguel Pereira, Cogitars, UK

Ensaios clínicos bayesianos – pequeno workshop baseado num ensaio muito conhecido

A estatística bayesiana tem sido cada vez mais utilizada em ensaios clínicos, oferecendo maior flexibilidade e eficiência no desenvolvimento de novos fármacos.

Neste seminário abordaremos este tópico utilizando como exemplo base num grande ensaio clínico muito conhecido mas que poucos sabem que utilizou métodos bayesianos. Vamos explorar em detalhe a metodologia utilizada no ensaio e em como é aplicável a outros ensaios. Será também abordado o tema de escolha do tipo de distribuições a priori e como escolher parâmetros de uma distribuição.

Joint seminar CEMAT and CEAUL

Europe/Lisbon
SASlab (6.4.29) Faculty of Sciences of the Universidade de Lisboa — Online

João Torrado Malato, CEAUL and IMM, University of Lisbon, Portugal and Warsaw University, Poland

Impact of misdiagnosis in case-control association studies: the case of myalgic encephalomyelitis/chronic fatigue syndrome

Misdiagnosis can occur when different case definitions are used by clinicians (relative misdiagnosis) or when failing the genuine diagnosis of another disease (misdiagnosis in a strict sense). In complex diseases, such as myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), this problem translates to a recurrent difficulty in reproducing research findings. To explore these effects, we simulated data from case-control studies under the assumption of misdiagnosis in a strict sense. We estimated the power to detect a genuine association between a potential causal factor and ME/CFS and demonstrated how current research studies may have suboptimal power. To address the implications of these findings, suggestions for how power can be improved are given and explained within the context of the disease.

Joint seminar CEMAT and CEAUL

Europe/Lisbon
SASlab (6.4.29) Faculty of Sciences of the Universidade de Lisboa — Online

Luis Gimeno-Sotelo, CEAUL and University of Vigo, Spain

Dependence modelling of extreme hydrological events in current and future climates

In this seminar, Dr. Luis Gimeno-Sotelo will provide an overview of his most recent advances on the extreme value analysis of the main hydrological extreme events (heavy rainfall and droughts) in terms of their main drivers. The most relevant statistical methods for non-stationary extreme value modelling will be presented, as well as a variety of methods from the copula theory to study bivariate extremes and conditional probabilities. He will explain the main applications of these statistical methodologies in the aforementioned environmental context, allowing for the identification of hotspot regions of high statistical dependence between the drivers and the hydrological extremes, as well as the analysis of the projected changes in the probabilities of occurrence of these extreme events in a global warming context.

Joint seminar CEMAT and CEAUL