Lecture notes (ST 790 M)

Lecture notes will be updated and available on the day of the class.

Lecture 1 (pdf file). Introduction to Spatial Data and models.
Syllabus (pdf)
Reading: Hierarchical Modeling and Analysis for Spatial Data. Pages 1 to 18.
(08/24/2006)
Lecture 2 (pdf file). Spatial processes, stationarity, and isotropy (studying lack of isotropy).
Reading: Hierarchical Modeling and Analysis for Spatial Data. Pages 21 to 29.
(08/29/2006)
Lecture 3 (pdf file). Examples of spatial processes, analysis of lack of stationarity and anisotropy for air pollution data: ozone,SO2, and nitric acid.
Reading: Hierarchical Modeling and Analysis for Spatial Data. Pages 21 to 29. R code to calculate geodesic distances. Code
Assignment: Exercise 6 in chapter 2.
(08/31/2006)
Lecture 4 (Notes, pdf file) Estimation of the covariance/variogram parameters and examples.
transparencies (pdf file). Transparencies used in class in case you want them (though the material in the transparencies is in the notes for Lecture 4!!).
Reading: Section 2.1 in the text book. (9/5/2006)
Winbugs tutorial. In Patterson #206 (Winbugs is not available in the computer lab). Please bring your laptops to class to learn how to use Winbugs for spatial analysis. You might want to load Winbugs first http://www.mrc-bsu.cam.ac.uk/bugs/ (9/7/2006)
We will continue with Lecture 4 (9/12/2006). Please stop by my office before the class (from 11 to 11:45) if you need help with the material covered so far.
Bayesian estimation of the spatial structure (notes from lecture 4). 9/14/06.
Lecture 5. Examples of likelihood estimation. Kriging spatial prediction and Bayesian Spatial prediction.
MLE examples (pdf file).
Spatial prediction (pdf file).
Lecture 6. Bayesian prediction and examples.
transparencies (pdf file). Transparencies used in class in case you want them for the Bayesian prediction (though the material in the transparencies is in the notes for Lecture 5!!). ASSIGNMENT: Exercise 3 in page 66. Due october 10.
(09/19/06)
R: Bayesian spatial estimation. dukelabnew.doc (notes in R) (9/21/06) TODAY THE CLASS WILL MEET IN THE COMPUTER LAB IN HARRELSON BUILDING, G100
Continuation with Bayesian prediction and MCMC, and starting areal models notes. We will do in class exercise 3 in page 66. 9/26/06
R: Bayesian spatial prediction. dukelabnew.doc (notes in R) (9/28/06) TODAY THE CLASS WILL MEET IN THE COMPUTER LAB IN HARRELSON BUILDING, G100
Lecture 7. Areal (lattice) data models. notes (pdf file) .
Reading: Chapter 3 in the textbook. 10/3/06.
Lecture 9: Bayesian Inference. notes ( pdf file) Reading: Chapter 4 in the textbook. (10/10/06- 10/17/06)

PLEASE REMEMBER EXERCISE 3 in page 66 is due today and also the Davis analysis.

Homework 1: Do a spatial analysis of the Davis dataset discussed in class: Plot the data; Fit a semivariogram function using WNLS and the likelihood; Interpret results and parameters. Give your interpretation of the dataset and you might compare it to other analysis done (see examples in the notes).
Due October 10.
The Davis data set used in class

o Hierarchical models notes 10/19/06

o Continuation of hierarchical models. Nonstationary models. Notes (pdf file). Reading: pages 149-159 in textbook. Example presented in class of a hierarchical nonstationary model. Example (ps) 10/24/06

o NOVEMBER 7—class in the computer LAB in Harreson. notes winbugs notes