Abstract #300114

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #300114
Activity Number: 243
Type: Contributed
Date/Time: Tuesday, August 5, 2003 : 10:30 AM to 12:20 PM
Sponsor: Section on Bayesian Stat. Sciences
Abstract - #300114
Title: A Bayesian Kriged-Kalman Model
Author(s): Sujit K. Sahu*+ and Kanti V. Mardia
Companies: University of Southampton and University of Leeds
Address: Faculty of Mathematical Studies, Highfield, Southampton, , SO17 1BJ, England
Keywords: spatial temporal modeling ; State Space Model ; Kriging ; Kalman filter ; Bayesian inference ; MCMC
Abstract:

A suitable model for analyzing spatio-temporal data needs to take into account variation in both space and time. The method of Kriging is a popular approach in spatial statistics which makes predictions for spatial data. Kalman filtering using dynamic models is often used to analyze temporal data. Recently, these approaches have been combined in a classical setup termed Kriged Kalman filter (KKF) model. In the combined model, the Kriging predictions dictate the optimal regression surface for incorporating spatial structure and the dynamic linear model framework is used to learn about temporal factors such as trends, autoregressive components, and cyclical variations. We consider a full Bayesian KKF and its MCMC implementation, in so doing we make several major advances in modeling. We work with a hierarchical model and compare this with the original nonhierarchical model. In addition, three new model components are incorporated: (1) a component modeling a plume or point source pollution, (2) a component providing the Fourier representation of seasonal effects, and (3) a regression component which takes care of time varying covariates.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2003 program

JSM 2003 For information, contact meetings@amstat.org or phone (703) 684-1221. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2003