PLEASE NOTE CHANGE OF LOCATION FOR WINTER QUARTER
M112 Alway Building, Medical Center
(next to the Dean's courtyard)
|DATE:||January 26, 2017|
|TIME:||1:30 - 2:50 pm|
|TITLE:||Parameter estimation in infectious disease epidemiology: Do you have any better ideas?|
|SPEAKER:||John Marshall, Assistant Professor in Residence, Biostatistics and Epidemiology
Division of Biostatistics and Epidemiology, School of Public Health
University of California, Berkeley
In mathematical epidemiology, the spread of infectious diseases through populations is often studied using systems of ordinary differential equations. In the simplest case, these describe the progression of people from susceptible to infected/infectious to recovered/removed (the SIR model). Variants of this model can be used to understand the population dynamics of HIV, TB, malaria, Ebola and Zika. Statistical methods are used to estimate the rates at which people transition from one state to the next. Some of the most popular methods for estimating these model parameters implement a Bayesian Markov Chain Monte Carlo (MCMC) approach in which: a) the posterior probability of specific parameter values is calculated given their prior distributions and observed data; and b) an MCMC algorithm is used to sample parameter values in order to approximate their posterior distributions. Sequential Monte Carlo methods are often used to approximate the likelihood for a stochastic model, and Approximate Bayesian computation is often used when the likelihood cannot be easily calculated. In this talk, I will demonstrate the application of these techniques to models of malaria transmission and novel mosquito control strategies; however, as a modeler first and foremost, rather than a statistician, I am interested to know, do you have any better ideas for parameter estimation?
Ghassan Hamra, Richard MacLehose, and David Richardson. Markov Chain Monte Carlo: an introduction for epidemiologists. Int J Epidemiol. 2013 Apr; 42(2): 627-634.
A Camacho, AJ Kucharski, S Funk, J Breman, P Piot, WJ Edmunds. Potential for large outbreaks of Ebola virus disease. Epidemics, Vol. 9, December 2014, pages 70-78.