Abstract.

In a crisis situation like the one we are passing through, contributions from science and academy may be helpful for understanding the evolution of the COVID-19 pandemic. In this brief talk, a cooperative experience in Spain will be introduced, focusing later on short-term prediction methods based on statistical tools and pointing out the problems that may arise due to data recording heterogeneity.

Abstract.

In the talk we will report about the different modeling and simulation approaches in Koblenz.

As part of the MOCOS consortium we are conducting stochastic microstructure simulations together with the universities of Kaiserslautern, Trier and Wroclaw. In Koblenz we are also doing parameter estimations based on SIR-type models. These models allow to simulate effects of schools openings, the influence of household structures onto the disease dynamics and to estimate the number of undetected cases. Since all this is ongoing and very active current research work, the special focus of the talk is not yet fixed.

Abstract.

Following the governmental intervention around March 12, after case numbers in Denmark had risen, an Expert Group was formed with participants from a number of Danish universities. The objective of the group is to establish a model-based foundation for advice to the health authorities, most importantly based on predictions of the load on the health care system in various scenarios. In this talk I will give an overview of the modelling frameworks used by the group and the analyses conducted. I will present some learnings from the work and point out areas where our limited knowledge about the disease is most critical.

Abstract

Simple assumptions on the spreading of a Covid-19 like virus lead to state a new epidemic model. Based on a nonlocal integrodifferential equation, it is able to capture key features of the current pandemic. While we leave the analytic well posedness theory to a future study, here we present numerical integrations reproducing qualitative patterns that have been observed in the last weeks. Within this new framework, various optimal control problems can be stated, possibly suggesting optimal health policies.

Joint work with M. Garavello, F. Marcellini, E. Rossi

Abstract.

Following the emergence of a novel coronavirus (SARS-CoV-2) and its spread outside of China, Europe has experienced large epidemics. In response, many European countries have implemented unprecedented non-pharmaceutical interventions including case isolation, the closure of schools and universities, banning of mass gatherings and/or public events, and most recently, wide-scale social distancing including local and national lockdowns.

In this talk we describe a semi-mechanistic Bayesian hierarchical model that infers the impact of these interventions and estimates the number of infections over time. We estimate parameters jointly for all countries in a single hierarchical model. Inference is performed in the probabilistic programming language Stan using an adaptive Hamiltonian Monte Carlo (HMC) sampler.