This short course is organized by the Mathematical Statistics and Data Science research group

Francesco Bartolucci

Francesco Bartolucci
(University of Perugia)
Bio Francesco Bartolucci is a Full Professor of Statistics in the Department of Economics at the University of Perugia (IT). He earned a Doctorate degree in “Statistical and Mathematical Methods for Economic and Social Research” in 1999. During 1997 and 2000 he visited the School of Mathematics and Statistics at the University of Sheffield (UK) and the Department of Statistics at the Penn State University (USA), respectively, where he was an instructor for graduate students. His research interests are: longitudinal and panel data, latent variable and mixture models, marginal models for categorical data, optimization, and Markov chain Monte Carlo algorithms. He served as principal investigator of the research project “Mixture and latent variable models for causal inference and analysis of socio-economic data”, which was funded by the Italian Government (Futuro in ricerca 2012) and of the project “Hidden Markov Models for Early Warning Systems” (PRIN 2022). He is an editor of Statistical Modelling: An International Journal.

Course Description

The short course aims to provide basic elements on latent variable models with a special focus on formulations based on latent variables having a discrete distribution. Among these formulations, the one leading to the hidden Markov model for time-series and longitudinal data will be discussed in detail, considering also the extended versions of this model involving individual covariates and random effects. Computational challenges involved in the estimation of these models by maximum likelihood and Bayesian methods will be discussed also with examples.

Detailed content

Lecture 1

  • Basic elements of latent variable models
  • Formulations based on discrete latent variables
  • Expectation-Maximization algorithm for maximum likelihood estimation
  • Bayesian estimation via augmented MCMC

Lecture 2

  • Basic hidden Markov models
  • Maximum likelihood estimation
  • Bayesian estimation

Lecture 3

  • Extended hidden Markov models with covariates
  • Inclusion of random effects
  • Applicative examples

References

Schedule

  • 26 October 2026, 16:30-18:30, room TBA
  • 27 October 2026 08:30-10:30, room A224
  • 28 October 2026, 08:30-10:30, room A223

Details

  • Venue: Polo Scientifico e Tecnologico F. Ferrari
  • Language: English
  • The participation is free. Please send an email to Prof. Veronica Vinciotti to confirm your participation.
  • For further information, please contact Prof. Veronica Vinciotti