Syllabus (click to download).
COURSE INFORMATION:
The course material will be based on a set of notes being prepared by the instructor. The recommended books are:
- Introduction to Geostatistics. P.K. Kitanidis. University Press. 1997.
(this is a basic level and helpful book
for the geostatistics part of this course. I recommend it ONLY
for students without any previous knowledge of spatial statistics
to get some background).
- Statistics for Spatial Data. Noel Cressie. Wiley & Sons. 1993.
(more complete
and more advance
level. It is a very good reference book but at an advance level.)
- Interpolation of Spatial Data. M. Stein. Springer, 1999.
(Very advance level, this is a good reference book for
spatial statistics in the spectral domain)
(These books are available in NCSU Central Bookstore and also on reserve at the main library).
The software used for
this course is SAS, and also Splus,
- There is a SAS Spatial Prediction notebook. SAS/STAT Technical Report:
Spatial Prediction Using the SAS System. It is available at the bookstore.
- There is a recommended Splus Spatial Statistics book.
S+Spatial Stats: User's manual. Mathsoft Inc.
It is available at the bookstore.
GRADING:
S/U
This course will cover a number of areas of spatial statistics applied to real, scientific and interesting problems. A
tentative list of more specific topics is as follows:
1. Introduction to spatial statistics.
2. Estimation of spatial correlations:
a. estimating variogram
b. fitting parametric models: Matern class
c. maximum likelihood estimation
d. restricted maximum likelihood
e. Bayesian procedures
f. MINQE estimation
3. Prediction and Interpolation (kriging):
a. Lagrange multiplier approach
b. conditional inference approach
c. Bayesian approach
d. predicting at multiple sites
e. frequentist corrections for unknown covariance structure
f. model misspecification in kriging
g. median polish kriging
4. Spectral domain:
a. Fourier Theory
b. Spectral Representation of a Spatial Process
c. Spectral Density
d. Periodogram
e. Increasing domain asymptotics
f. Infill asymptotics
5. Spatial-temporal processes.
6. Design of spatial networks.
7. Nonstationary spatial processes.
8. Lattice data: Mantel test.
Students will learn how to use existing software, the emphasis of
the course is to learn
the methodology needed to do research on spatial statistics
and to analyze
real data from the environmental, geological and
agricultural sciences.
There will be a project (a spatial analysis using
real data provided by Dr Fuentes or selected
by the students), a (in-class) midtern and a final take home.
|