Reference : Inference in a stochastic SEIR model using Sequential Monte Carlo methods
Scientific congresses and symposiums : Paper published in a book
Physical, chemical, mathematical & earth Sciences : Mathematics
http://hdl.handle.net/2268/202086
Inference in a stochastic SEIR model using Sequential Monte Carlo methods
English
Bonou, Wilfried mailto [Université de Liège > Faculté des sciences sociales > Méthodes quantitatives en sciences sociales >]
LAMBERT, Philippe mailto [Université de Liège - ULiège > Faculté des Sciences Sociales > Méthodes Quantitatives en Sciences Sociales > >]
Aug-2016
The 37th Annual Conference of the International Society for Clinical Biostatistics (ISCB): Book of Abstracts. Birmingham, UK, 21-25 August 2016
Yes
International
The 37th Annual Conference of the International Society for Clinical Biostatistics (ISCB)
21-25 August 2016.
Birmingham
UK
[en] Stochastic Dynamic Epidemic Model ; State Space Models ; Sequential Monte Carlo
[en] Many biological, physical, chemical, economic, and social phenomena are dynamic and are modeled using (systems of) Ordinary Differential Equations (ODEs). But a more realistic way to describe these dynamics relies on Discrete Time Markov Chains (DTMC). There is a growing interest in the development of Bayesian statistical methods to infer on the parameters in such dynamic models, particularly those defining epidemic spread, by combining prior information with experimental or observational data. Our proposal aims to explore the merits of the Bayesian Optimal Filtering technique in the estimation of the parameters of a stochastic SEIR (S = Susceptible, E = Exposed, I = Infectious, R = Removed) epidemic model. State Space Models (SSMs) are used to describe the epidemic dynamic. The unknown static parameters are estimated using a combination of Sequential Monte Carlo techniques with a Markov Chain Monte Carlo algorithm, . . .
http://hdl.handle.net/2268/202086

There is no file associated with this reference.

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.