Theory of Continuous Space and Space-Time Processes - Fall 2009

Course Description:
This course is intended to provide a strong theoretical foundation for space and space-time processes over continuous domains. Topics will include continuous parameter stochastic process theory; spectral methods; spatial asymptotics; nonstationary spatial modeling; dynamic models and spatial time series; nonseparable space-time models; spatial design; space-time data fusion; low rank representations; nonparametric spatial methods; topics in shape analysis.

Lectures

August 27 - Alan Gelfand, A Short Course on Multivariate Spatial Process Modeling ;
First Look at Multivariate Processes
September 3 - Alan Gelfand, Coregionalization and Other Constructions
September 10 - Alan Gelfand, Handling Large Space and Space time Datasets
September 17 - No class due to Opening Workshop
September 24 - Montse Fuentes, Spectral Methods for Spatial Data
October 1 - Sakis Micheas, Theory for Continuous Spatial Processes or Shape Analysis
October 8 - Kate Calder, Space-time Statistical Modeling
October 15 - Veronica Berrocal, Data Fusion Approaches for Space and Space Time Data
October 22 - Alan Gelfand, Nonparametric Modeling for Spatial Data
October 29 - Katja Ickstadt, Point Processes
November 5 - Dongchu Sun, Prior Specifications for Space and Space-time data
November 12 - Murali Haran, Computer Models (Using Gaussian Processes)
November 19 - Student Presentations

Student presentations may be applied, say fitting a model discussed during the course to an appropriate dataset, or a presentation of some technical work you have done, or a discussion of a few important papers. All presentations will be done as posters given on 19 Nov. I will work out the logistics between now and then. Your proposal for your poster must be submitted to me by no later than 8 Oct. You are encouraged to develop and discuss your presentation based upon the course lectures and to speak with the various visitors to the SAMSI spatial program to help you do this.

Alan Gelfand

Nonstationary Multivariate Process Modeling through Spatially Varying Coregionalization

Supplemental Materials:
A Spatial Statistical Analysis of Tumor Growth
A Bayesian Hierarchical Nonoverlapping Random Disk Growth Model
Statistics of the Boolean Model: From the Estimation of Means to the Estimation of Distributions
Second-order analysis of inhomegeous spatio-temporal point process data

Registration for this course is being processed through your university.

Course Numbers:
Duke: 294-03
NCSU: MA/ST 810.002
UNC: MATH 891-003/STOR 960-001