Prof. Dr. Leonhard Held
Building a Statistical Model: The Endemic-Epidemic Modelling Framework
Was |
|
---|---|
Wann |
03.11.2017 von 14:00 bis 15:00 |
Wo | Eckerstraße 1, Raum 404, 4. OG |
Termin übernehmen |
vCal iCal |
Novel statistical methods arise as a result of contact with data combined with concrete research questions. Statistical modelling has been described as the art to find a suitable mathematical framework based on probability distributions that allows the questions at hand to be answered or new ones to be posed (Davison, 2003, Statistical Models, Cambridge University Press). To this end, it is often necessary to find a pragmatic balance between parsimony and complexity.
In this talk I will review the development of the endemic-epidemic modeling framework for multivariate time series of infectious disease counts (Held et al, Stat Med, 36, 3443-3460, 2017 and references therein). The formulation is built upon an additive decomposition of disease incidence into an endemic and an epidemic component. The endemic component may represent seasonal and climatic variation, heterogeneity in population numbers and other socio-demographic characteristics. The epidemic component describes the force of previously infected individuals through spatio-temporal or social interaction. I will discuss how research questions from epidemiology and public health have been coupled with mathematical theory, statistical methodology and the available surveillance data to study and to predict the spread of infectious diseases.