Selection of Parametric Selection of Parametric Networks in PRRS using ILMs, and Linearization of ILMs for Large-Scale Spatial Infectious Disease Systems

Dr. Grace Pui Sze Kwong Biostatistics Consultant Faculty of Veterinary Medicine University of Calgary Abstract: Individual-level models (ILMs) for infectious diseases, fitted in a Bayesian MCMC framework, are an intuitive and flexible class of models that can take into account population heterogeneity via various individual-level covariates. This talk addresses two separate issues. The first part of this talk is to identify relative importance of risk factors for the spread of the genotype 1-18-4 PRRS virus from monitoring data in southern Ontario using ILMs. Porcine reproductive and respiratory syndrome (PRRS) has a worldwide distribution. This economically important endemic disease causes reproductive failure in breeding stock and respiratory tract illness in young pigs. Here, we explore networks through which resources are obtained or delivered, as well as the ownership structure of herds, and identify factors that may be contributing to high risk of infection. ILMs containing a geometric distance kernel to account for geographic heterogeneity provide a natural way to model the spatial spread of many diseases. However, in even only moderately large populations, the likelihood calculations required can be prohibitively time consuming. It is possible to speed up the computation via a technique which makes use a linearized distance kernel. Here in the second part of the talk, we examine some methods of carrying out this linearization and compare the performances of these methods.

More information at http://www.ucalgary.ca/events/calendar/selection-parametric-selection-parametric-networks-prrs-using-ilms-and-linearization-ilms