All Times EDT
Virtual
WITHDRAWN - Does Bayesian Kriging Reduce Exposure Measurement Error When Applied to Weather Data? (309565)
Nina Cesare, Boston UniversityLeah Forman, Boston University
Kevin Lane, Boston University
*Amanda Elizabeth Norton, Boston University
Keith Spangler, Boston University
Keywords: Weather, Kriging, Interpolation, Spatial Analysis, Spatial Statistics, GIS
Weather data is collected via monitors which represent points on a map. This presents an issue as there are many areas where there is not monitor coverage and therefore the data must be interpolated. Spatial interpolation creates exposure measurement error, as models are being used to estimate values which are dependent on complex systems. This analysis will use three common interpolation methods, inverse distance weighting (IDW), ordinary kriging (OK), and Empirical Bayesian Kriging (EBK) to estimate temperature across the United States. We will us a 20% out of sample cross validation method to assess the accuracy and bias of these interpolation methods EBK generates kriging model parameters automatically via simulation, we hypothesize t that it is a more accurate interpolation method than IDW or OK. These findings may guide researchers in choosing interpolation strategies for a variety of spatial processes.