Lynnette M. Smith, of the Department of Biostatistics at the University of Nebraska Medical Center, will present:
Abstract: In Public Health we strive to predict disease incidence or mortality rates for a particular area, such as state, county, or zip code. At an aggregate level, we can predict spatially correlated count data using a Generalized Linear Mixed Models (GLMM) framework with a Poisson outcome variable with an auxiliary variable in a bivariate relationship. A cokriging structure is applied in a GLMM model setting which includes a Poisson outcome variable and an auxiliary variable with a named distribution, where both the outcome variable and auxiliary variable are spatially correlated and correlated with each other. This methodology is examined in a real data setting with an application in public health, predicting West Nile virus incidence at the county level. Environmental auxiliary variables considered in the West Nile virus prediction are counts of infected mosquitos and birds and percent irrigated farmland. To evaluate the prediction performance, cross-validation is used, comparing predicted vs. actual incidence.
A social tea will be held at 3:00 p.m. in the CCBR Ballroom. All are Welcome.