Home / News / Mini-Workshop on Epidemiological Modeling

Mini-Workshop on Epidemiological Modeling

The Modeling and Applications Research Group is inviting everyone to a Mini-workshop on Epidemiological Modeling to be held on Tuesday, October 24, 2023, 1:00 – 4:00 PM at MBAN 103.  The workshop will be conducted by Dr. Lorenzo Rimella, Senior Research Associate, Department of Mathematics and Statistics, Lancaster University, UK.

The workshop aims to provide participants with a comprehensive understanding of compartmental models for epidemiological modelling and associated techniques for likelihood computation and parameter inference. Real-world data will be presented throughout the workshop to illustrate the challenges associated with these models.

Topics to be covered includes:

  • Compartmental models as hidden Markov models (HMM)
    • Likelihood computation in HMM and issues
    • Likelihood approximation through SMC
    • Parameters inference
  • Parametric Approximation in compartmental models
    • Kalman filter
    • Beyond the Gaussian approximation: The Poisson approximate likelihood (PAL) and the Multinomial approximate likelihood (MAL)
    • Dealing with overdispersion
  • Individual-based model
    • Definition, motivation, and challenges
    • Likelihood computation and inference

About the speaker: Lorenzo Rimella completed a “B.Sc. in Mathematics for Finance and Insurance” and an “M.Sc. in Stochastics and Data Science” from UNITO (Università degli studi di Torino). He subsequently pursued a Ph.D. in high-dimensional Statistics at the University of Bristol under the supervision of Prof. Nick Whiteley. During this Ph.D. studies, he also served as an enrichment student at The Alan Turing Institute, the UK’s national institute for data science and artificial intelligence, located in London. Currently, he holds the position of senior research associate at Lancaster University, where he collaborates with Paul Fearnhead and Chris Jewell on epidemiological modelling. His research interests cover multiple aspects of computational statistics, and can be grouped under the general domain of “high-dimensional filtering”, with epidemiology as the primary real-world application.