My research focuses on Bayesian models for longitudinal data and Bayesian mixture models for community detection problems. I have a deep passion for statistics in all its facets, from theoretical modeling to computational techniques, with a particular emphasis on applications to real-world data. I am especially interested in data derived from epidemiological, biological, and medical contexts, aiming to bridge the gap between theoretical research and practical applications.
SHORT BIO
I first approached the research supported by my M.Sc. thesis supervisors, Professors Enrico Bibbona and Gianluca Mastrantonio (Politecnico di Torino). With them, I had the chance to face the first challenges in the pharmacokinetic research field, with a M.Sc. thesis devoted to develop algorithms to infer hidden Markov models driven by diffusions.
During my Ph.D. experience, supervised by Professor Mauro Gasparini, I got in touch with several real datasets and the relative questions of interest. Starting from the dataset of interest, I created hierarchical mixed models to be realistic enough to describe the problem, but simple enough to allow identifiability during estimation procedures. I got in touch with many techniques with the intent of making inference, from frequentist algorithms such as Stochastic Approximation Expectation Maximization (SAEM) to hybrid versions such as SAEM-MCMC, and, finally, the more standard MCMC. In the latter part of my Ph.D. I also got in touch with Causal Inference, thanks to Prof. Stijn Vansteelandt and his research group. This changed a little bit my way of looking at target questions and classical inference techniques.
All these works are part of my Ph.D thesis, entitled Statistical methods for longitudinal medical data with applications, which I succesfully defended on May 2024.
During the Ph.D., I also had the great opportunity to collaborate with researchers of GSK Vaccine and with clinicians from various medical departments. This allowed me to learn much more about real problems that can arise in practical applications, behind the simplified theoretical framework on which I focused during my studies. This taught me how to collaborate with non-statisticians, which I really see as one of the most important results out of these years of doctoral experience.
I am currently working as a Postdoctoral Fellow at the Department of Economics, Social Studies, Applied Mathematics and Statistics (Torino, Italy), with a fellowship on Bayesian mixture modeling for community detection in networks, under the supervision of Professor Pierpaolo De Blasi.