Upcoming GRBIO seminars

Sessió

Maximum likelihood estimation of a regression model with two interval-censored covariates

Abstract

Interval-censored covariates arise in many biomedical studies when the exact value of a covariate cannot be observed but is known to lie within an interval. Regression analysis with an interval-censored covariate is already challenging, and allowing for two such covariates adds a further layer of complexity because their dependence must also be accounted for. The setting considered is a parametric survival regression model with a right-censored response and two interval-censored covariates. The joint distribution of the covariates is modeled through a copula, while the marginal distributions are left unspecified and estimated nonparametrically. We discuss identifiability of the resulting semiparametric model and its likelihood-based estimation.

Biosketch

Andrea Toloba is graduated in Mathematics at the University of Barcelona (2019) and pursued the Master's degree in Statistics and Operations Research (MESIO UPC-UB). During her MSc studies, she worked as statistician in the Epidemiology and public health Programme for Cardiovascular diseases at the Hospital del Mar Medical Research Institute (IMIM). Since September 2022, she is a PhD Student in the Statistics and Operations Research Programme at the Polytechnic University of Catalonia (UPC), under the supervision of Prof. Guadalupe Gómez Melis and Prof. Klaus Langohr. Her main research topic focuses on interval-censored data.

Sessió

A new approach for generating spatial synthetic data using archetypes

Abstract

The generation of synthetic data is an increasingly relevant topic in modern statistics, serving a wide range of purposes, from facilitating data sharing to augmenting datasets. Within this context, this seminar introduces a novel framework specifically designed for generating georeferenced synthetic data. By leveraging Archetypal Analysis (AA) to identify extreme profiles (archetypes) within a dataset, we propose a method to construct realistic synthetic datasets that aim to replicate the spatial structure and dependence of the original observations. The presentation will outline the proposed methodology and the computational strategies adopted for the generation process. Finally, we will discuss preliminary results and future directions of this research.

Biosketch

Federica Alberti graduated in Economics and Data Analysis (2022) and obtained a Master's degree in Data Science for Economics and Management (2024) at the University of Brescia (Italy). Since November 2024, she has been a PhD student in Analytics for Economics and Management, within the Statistics and Data Science area, at the same institution, under the supervision of Prof. Marica Manisera and Prof. Paola Zuccolotto. In 2026, she completed a four-month visiting stay at the Universidad de Castilla-La Mancha (UCLM) and is currently a visiting PhD student at the Universitat Politècnica de Catalunya (UPC). Her research focuses on spatial statistics and related topics.

Sessió

Landmark Multi-State Modeling for Complex Disease Trajectories

Abstract

Analyzing complex disease trajectories using multi-state models (MSMs) requires handling both complex event structures and the computational demands of large-scale registries. We introduce an MSM framework extending beyond simple illness-death structures to model time-to-recurrence (relapse or reinfection) by integrating data from initial events. We address two challenges: defining recurrence at individualized landmark constraints, and developing optimized algorithms to mitigate the severe computational burden imposed by large-scale datasets.

Biosketch

Guadalupe Gómez Melis is Professor at the Universitat Politècnica de Catalunya and of the Reial Acadèmia de Ciències i Arts de Barcelona. She is member of the Executive Board of the International Biometric Society (IBS) and elected member of the Council of IRIS, the Institute for Research and Innovation in Health by UPC. Past Member of the UPC Ethics Committee, Past President of the Consell Català d’Estadística, Past-Elected European Representative of Caucus for Women in Statistics, Past-Council Elected Member of International Biometric Society, Past-Member Expert Committee of the Acción Matemática contra el coronavirus del Comité Española de Matemáticas. Former Vice-Dean in the School of Mathematic and Statistics, coordinator of the Master in Statistics and Operations Research UPC-UB, Coordinator of the Interuniversity Doctorate in Bioinformatics and Coordinator of the Doctorate in Statistics and Operations Research. She graduated in Mathematics at the Universitat de Barcelona (UB). Her scientific career started in the Statistics Department of Columbia University, NY, where under the direction of Prof. John Van Ryzin, obtained her PhD. After a short stay as Assistant Professor in the Statistics Department at Ohio State University, she joined the Statistics and Operations Research Department at the Universitat Politècnica de Catalunya-BarcelonaTECH (UPC) where she became Full Professor in 2005. She has been visiting scientist, among others, at: Biostatistics Department at Harvard University, Boston (USA) where she collaborated with Prof. Steve Lagakos; Short Term Research Collaborator at Oxford University Clinical Research Unit, Ho Chi Minh City (Vietnam) collaborating with Dr. Ronald Geskus and MD Anderson Cancer Center (MDACC), the University of Texas, Houston (USA) where she collaborated with Prof. Yu Shen. Her research commitment has always been combined with a determined leadership and service. Under her leadership the Research Group in Biostatistics and Bioinformatics (GRBIO) (SGR2014-464, SGR2017-622, SGR2021-01421) emerged. GRBIO addresses challenging problems raised in Life and Health Sciences in the frontier of both disciplines. She has been the PI of more than 20 funded projects of the Ministerio de Ciencia e Innovación since 1992. Her main research interests are in developing methods for Survival Analysis and Clinical Trials, with an unequivocal interdisciplinary flavor, focusing especially on cancer, HIV-AIDS and on Influenza and COVID-19, as well as Liquid Chromatography. She has collaborated with more than 50 researchers and publishes in high impact journals such as Statistics in Medicine, Biostatistics, Biometrika, Biometrics, Statistical Methods in Medical Research, European Journal of Cancer, Frontiers Immunology, Journal of Clinical Epidemiology.

Sessió

Títol

Abstract

Descripció

Biosketch

Leire Garmendia graduated in Mathematics at the University of Basque Country (UPV/EHU) and obtained her MSc in Statistics and Operations Research at the Universitat Politècnica de Catalunya (UPC). Currently, she is a PhD student in Statistics at the University of Basque Country (UPV/EHU).