Talks

Name: Ana Luísa Abreu Gonçalves

Title: Maximum likelihood estimation of a GLM with hyperbolic link function – an application on agronomic data.

Name: Ana Sofia Tedim

Title: Does vaccination impact COVID-19 transmission?

Abstract: Vaccination has been used as a strategy to mitigate the effects of Covid 19 pandemic. Here, we analyzed epidemiological data from Aveiro region and proposed a descriptive model for the number of secondary cases (new infections attributed to one individual) generated by an infected person. Our goal is to evaluate the effect of vaccination on this variable.

Name: Juliana Reis

Title: An overview of statistical analysis of adaptive designs in clinical trials

Abstract: In recent years, the use of adaptive design methods in clinical trials has become popular due to their flexibility and efficiency. An adaptive design is defined as a clinical trial design that allows prospectively planned modifications to one or more aspects of the design based on accumulating data from subjects in the trial. Group sequential designs are the most commonly used design to facilitate the conduct of interim analysis in an adaptive clinical trial. The objective is to present a synthesis of the sequential group methods most used in adaptive designs in different situations, illustrating the application of different methodologies in practical real situations.

Name: Leonor Carvalho Rodrigues

Title: Stable variable selection in penalized regression models: an application to high dimensional genomic data - Alzheimer’s Disease

Abstract: Alzheimer’s disease (AD) is a complex disorder caused by a combination of environmental and genetic factors. Since the majority of these diseases have not an available treatment already, it is important to make an early and accurate detection to prevent disease progression. One of the main goals of modern genetics has been unraveling the genetic background of common complex disorders. The challenge in finding a plausible method to apply in genomic data is due to its high dimensionality. Penalization techniques have already been applied in the context of Genome-Wide Association Studies (GWAS). However, some instability is associated with these techniques. In this way, the main goal of this work is to find a consistent method that identifies a correlation between some Single Nucleotide Polymorphisms (SNPs) and AD in a structure with a huge number of potential predictor variables. To achieve this, penalized regression methods (LASSO and Elastic-net) were applied in a combined way with methods based on Akaike's Information Criterion (AIC) to evaluate the importance of potential predictors.

Name: Márcia L. Silva

Title: Prey-predator model with constant-effort harvesting

Abstract: We study a prey-predator model based on the classical Lotka-Volterra system with Leslie-Gower and Holling IV schemes and a constant-effort harvesting. Our goal is twofold: to present the model proposed by Cheng and Zhang in 2021, introducing some changes and corrections; to analyze the number and type of equilibrium points for the modified model. We end by proving the stability of the meaningful equilibrium point, according to the distribution of the eigenvalues.

Name: Vanusa Rocha

Title: Has mortality due to COVID 19 been decreasing over time?

Abstract: COVID 19 is an infectious disease caused by SARS-CoV-2, first identified in December 2019 in Wuhan City, China. The initial outbreak eventually led to a global pandemic causing many deaths around the world. The aim of this study is to evaluate COVID 19 mortality in Portugal, perceive the behavior over time and identify associated risk factors. The database studied contains epidemiological surveillance data for the disease from March 3, 2020 to July 11, 2021. Potential risk factors for death COVID-19 were analyzed by a multivariable logistic regression and Cox proportional hazard regression.

Mortality from COVID 19 has been decreasing over time and, in this work, we will discuss possible reasons associated with that decrease.