Bayesian Statistics Seminar
North Carolina State University
presents
Dr. Sudipto Banerjee
University of Minnesota
"Hierarchical spatial models for predicting forest variables over large heterogeneous domains"
ABSTRACT
We are interested in predicting one or more continuous forest variables (e.g., biomass, volume, age) at a fine resolution (e.g., pixel-level) across a specified domain. Given a definition of forest/non-forest, this prediction is typically a two step process. The first step predicts which locations are forested. The second step predicts the value of the variable for only those forested locations. Rarely is the forest/non-forest predicted without error. However, the uncertainty in this prediction is typically not propagated through to the subsequent prediction of the forest variable of interest. Failure to acknowledge this error can result in biased and perhaps falsely precise estimates. In response to this problem, we offer a modeling framework that will allow propagation of this uncertainty. Here we envision two latent processes generating the data. The first is a continuous spatial process while the second is a binary spatial process. We assume that the processes are independent of each other. The continuous spatial process controls the spatial association structure of the forest variable of interest, while a binary process indicates presence of a ``measurable'' quantity at a given location. Finally, we explore the use of a predictive process for both the continuous and binary processes to reduce the dimensionality of the data and ease the computational burden. The proposed models are motivated using georeferenced National Forest Inventory (NFI) data and coinciding remotely sensed predictor variables.
This is joint work with Andrew O. Finley (Department of Forestry and Geography, Michigan State University)
Thursday, November, 19, 2009
4:00 - 5:00 pm
450 Riddick Hall