Invited Speakers
Prof. Teresa A. Oliveira
Universidade Aberta
Teresa A. Oliveira is Associate Professor with Habilitation in Mathematics-Statistics, at DCeT-UAb, in Lisbon, Portugal. She obtained MSc and doctoral degrees in Statistics and Operations Research at the University of Lisbon and is a senior member of the CEAUL, http://ceaul.org/o-centro/membros-ceaul/. Her research interests include Experimental Design, Statistical Quality Control, Risk Analysis, Statistical Modeling, Computational Statistics, Stochastic Models, Data Analysis, and e-Learning Methodologies. She actively participated in many Erasmus+ Bilateral and Teaching Programs and was selected by the National Agency as an evaluator for Erasmus+ during 2021/2027. She is the Chair of the ISI-CRA International Statistical Institute - Committee on Risk Analysis, and was appointed as a founding member of the ISI Working Group on Data Science. With extensive editorial experience, she is Associate Editor of JAS - Journal of Applied Statistics and MPS - Mathematical Population Studies, as well as of other prestigious Journal Editorial Boards. She has publications in several peer-reviewed international and national papers, books, special issues, and proceedings, and her main research results were presented in many conferences worldwide. Her contributions to Mathematics-Statistics continue to inspire new research and innovation, which is clearly reflected in theses and dissertations under her supervision, in various courses, and in several universities.
Prof.
Clara Cordeiro
University of the Algarve
and CEAUL
Clara Cordeiro holds a PhD in Mathematics and Statistics from the Technical University of Lisbon, specialising in forecasting time series with Bootstrap. She is an Assistant Professor in the Department of Mathematics at the University of the Algarve. Clara is also a member of the Research and Development unit - Centre of Statistics and Applications (CEAUL) at the University of Lisbon. Her scientific interests are bootstrap for dependent data, exponential smoothing methods, and forecasting time series. Additionally, she has developed statistical models and algorithms to address scientific inquiries across various scientific domains. Recently, her research has been driven by collaborative projects with non-academic partners.
Speech Title: "Time Series Forecasting with Bootstrap Magic"
Abstract: Predicting the future is, and will always be, a huge challenge. Many researchers, using different forecasting methods, have tried to develop and update their knowledge to make their procedures more competitive than those already existing. Let us remember the phrase by George E. P. Box: 'Statisticians, like artists, have the bad habit of falling in love with their models.' In my previous studies, the combination of exponential smoothing methods and the bootstrap methodology has proven to be a promising association. Both are versatile and very popular procedures used in various research fields. The effectiveness of such a partnership will be illustrated through an empirical study demonstrating the utility of the bootstrap in forecasting time series. Acknowledgements: This work is partially financed by national funds through FCT – Fundação para a Ciência e a Tecnologia under the project UIDB/00006/2020. DOI: 10.54499/UIDB/00006/2020 (https://doi.org/10.54499/UIDB/00006/2020)
Asst.
Prof. Miguel Fonseca
NOVA University Lisbon
Miguel Fonseca is a member of the NOVA Math research center and holds a position of assistant professor in the department of Mathematics of the NOVA School of Science and Technology. His main research interests are multivariate statistics, statistical inference and machine learning. With a PhD in Mathematics, specialty of Statistics, he has vast activity in the academia with publications in statistics and applications in biostatistics and other areas. He also has experience in data analysis and software programming, both for public and private sector.
Speech Title: "Singular Covariance Matrix Models with Random Effects"
Abstract: In this paper, we consider multivariate mixed models with a dispersion matrix that is singular. We derive estimates for both fixed and random effects using maximum likelihood, considering several conditions restricting the fixed effects of the model. The efficiency of the estimators is ascertained with simulation studies.
.