|
Activity Number:
|
527
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #303681 |
|
Title:
|
Malaria Transmission Dynamics: A Formal Comparison of Rival Hypotheses
|
|
Author(s):
|
Anindya Bhadra*+ and Edward Ionides
|
|
Companies:
|
University of Michigan and University of Michigan
|
|
Address:
|
439 West Hall, Ann Arbor, MI, 48109,
|
|
Keywords:
|
malaria ; iterated filtering ; mechanistic modeling ; diasease dynamics
|
|
Abstract:
|
Malaria is a major cause of death in developing countries and a lot of attention has been focused to the statistical modeling of the disease dynamics. A model that is scientifically plausible tends to be too complicated and poses significant identifiability issues to the statistical fit. Most analyses so far use simpler non-mechanistic models that are hard to interpret scientifically. We describe a novel mechanistic coupled stochastic differential equation model of malaria driven by Levy noise - complex enough to capture the essential dynamics while maintaining the feasibility of data analysis. Our results, a comparison of several nested statistical models fitted by the maximum-likelihood method, provide a systematic framework for the model selection and prediction problems. Our use of the iterated filtering technique allows us to fit and compare models that were intractable so far.
|
