Inference for Ecological Dynamical Systems A Case Study of Two Endemic Diseases.pdfVIP

Hindawi Publishing Corporation Computational and Mathematical Methods in Medicine Volume 2012, Article ID 390694, 13 pages doi:10.1155/2012/390694 Research Article Inference for Ecological Dynamical Systems: A Case Study of Two Endemic Diseases Daniel A. Vasco Department of Biology, Duke University, Box 90338, Durham, NC 27708, USA Correspondence should be addressed to Daniel A. Vasco, vasco.daniel.a@ Received 2 September 2011; Revised 19 November 2011; Accepted 21 November 2011 Academic Editor: Vikas Rai Copyright © 2012 Daniel A. Vasco. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. A Bayesian Markov chain Monte Carlo method is used to infer parameters for an open stochastic epidemiological model: the Markovian susceptible-infected-recovered (SIR) model, which is suitable for modeling and simulating recurrent epidemics. This allows exploring two major problems of inference appearing in many mechanistic population models. First, trajectories of these processes are often only partly observed. For example, during an epidemic the transmission process is only partly observable: one cannot record infection times. Therefore, one only records cases (infections) as the observations. As a result some means of imputing or reconstructing individuals in the susceptible cases class must be accomplished. Second, the official reporting of observations (cases in epidemiology) is typically done not as they are actually recorded but at some temporal interval over which they have been aggregated. To address these issues, this paper investigates the following problems. Parameter inference for a perfectly sample