Principal Lecturer
·
Noel
Cressie (Ohio
State University)
Dr. Cressie will present lectures that concentrate
on spatial processes with continuous index. The models used to
describe spatial dependence are variogram- or
covariance-function-based, commonly called geostatistical
models. When embedded in a hierarchical model, geostatistical
models can provide statistically optimal spatial predictors that address
important science problems, which include spatial mapping: (1) on river
networks; (2) in the presence of location error; (3) for exceedances and exceedance
regions; (4) for massive datasets remotely sensed over the globe.
Spatio-temporal mapping, where both past and current data are used to
filter and forecast a spatio-temporal process,
will be presented in the context of remote sensing. The emphasis of these
lectures will be on hierarchical statistical modeling for processes with
continuous spatial index. Opportunities will be taken to put the
formulation in a Bayesian context and to point out connections with
spatial lattice processes.
Invited Speakers
·
Alan
Gelfand (Duke
University), "Stochastic Space-time Modelling
Using Differential Equations"
·
Marc Genton (Texas
A&M University),
"Modeling and Testing Properties of
Space-time Covariance Functions"
·
Richard
Smith (University of North Carolina--Chapel
Hill), "Extreme Precipitation
Trends Over the Continental United States"
·
Jay Ver Hoef (National Marine
Mammal Laboratory), "Space-Time
Zero-Inflated Count Models of Harbor Seals"
·
Christopher
Wikle (University
of Missouri), "Nonlinear Spatio-Temporal Dynamic
Models"
·
Jun Zhu (University of Wisconsin--Madison), "Markov
Chain Monte Carlo for a Spatial-Temporal Autologistic
Regression Model"
·
James Zidek (University
of British Columbia),
"Using a Multivariate Approach to
Model Univariate Environmental Space Time”
Processes”
·
Sujit Ghosh (North
Carolina State
University), “A Class of Kernel-Based Conditionally Autoregressive
Models for Spatial Data”
·
Victor De
Oliveira (The University of Texas at San
Antonio), “Objective
Bayesian Analysis of Spatial Data With Measurement Error”
Contributed and Poster Sessions: Authors are invited to make a contribution to
the conference, by submitting an abstract before January 31, 2007. Send your abstract by email to Joon Jin Song. Please indicate your
preference in regard to having an oral or poster presentation.
