Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

DY: Fachverband Dynamik und Statistische Physik

DY 7: Modeling and Data Analysis

DY 7.6: Vortrag

Montag, 12. März 2018, 11:15–11:30, BH-N 128

Profile likelihood based analyses of infectious disease models — •Christian Tönsing¹, Jens Timmer^1,2,3,4, and Clemens Kreutz^1,2,4 — ¹Institute of Physics, University of Freiburg, Freiburg im Breisgau, Germany — ²Freiburg Center of Data Analysis and Modeling (FDM), University of Freiburg, Freiburg im Breisgau, Germany — ³BIOSS Centre for Biological Signalling Studies, University of Freiburg, Freiburg im Breisgau, Germany — ⁴Center for Biosystems Analysis (ZBSA), University of Freiburg, Freiburg im Breisgau, Germany

Ordinary differential equation (ODE) models are frequently applied to describe the dynamics of epidemics. In this work, we use such models of infectious diseases for the estimation of a priori unknown model parameters and their uncertainties from the information contained in recorded data of infected individuals. A deterministic multistart optimization approach is applied for parameter estimation. Moreover, we introduce profile likelihood-based uncertainty analyses and check the identifiability of a simple SIR model with data from an influenza outbreak at an English boarding school in 1978. Furthermore, a complex ODE model for vector-borne diseases with data from the Zika virus (ZIKV) outbreak in Colombia in 2015/16 is used for data-based model reduction utilizing likelihood profiles.