You are here


Publication Date: 

Gaia Bertarelli has joined the Institute of Management of Sant’Anna School and EMbeDS as an Assistant Professor (RDT-A) in Social Statistics working in the Management and Healthcare Laboratory (MeS). Previously, she worked as a research fellow at the Universities of Pisa and Perugia, and as a statistician at the National Cancer Institute and at the Department of Anaesthesia and Intensive Care, Vita-Salute San Raffaele University, Milano.  

She was awarded a Ph.D. in Statistics from the University of Milano Bicocca in 2015.  Prior to her doctoral studies, she graduated in Mathematics (University of Milano Bicocca) and received a MSc. in Biostatistics and Experimental Statistics (University of Milano Bicocca). 

Gaia’s studies were mainly devoted to statistical methods for Official Statistics. In her Ph.D. thesis, she focused on latent variable models for aggregate data in disease mapping and small area estimation. Since 2015, Gaia has been collaborating with the Italian National Statistical Institute through several research projects on the integration of data coming from different data sources and on small area estimation. Her research fields include statistical methods for unplanned domains, robust statistics, latent variable models and multidimensional composite indicators. The fields of application of her research concern healthcare, gender differences, and the multidimensional aspects of poverty. 

Gaia has been a peer reviewer for journals such as Biometrical Journal, Journal of Official Statistics, Journal of the Royal Statistical Society Series A (Statistics in Society), Statistical Methods and Applications, Nutrients. In 2020, she chaired the section y-SIS -- the under 35 statisticians’ group of SIS (Società Italiana di Statistica). 

What are Gaia’s plans for this new chapter of her academic career, in connection with the objectives of EMbeDS? "The biggest challenge of my research will be to develop and test new statistical methods useful for the healthcare system, which allow one to integrate data from multiple sources and to estimate multidimensional composite indicators in unplanned domains, where the sample size is small and the reliability of traditional methods is hindered."