GEOS 597e
Spatiotemporal Data Analysis Workshop
Prework 6: EOF rotation
Last updated 10/11/06. To be
completed prior to class session Weds., Oct. 11th.
Introduction:
This week we'll analyze our
results from Homework 4 and 5. How does varimax rotation affect
the leading patterns we've extracted from the the global SST field?
Reading:
- Wilks,
2006, Ch. 11.5: Rotation of the Eigenvectors. In: Statistical Methods in the Atmospheric
Sciences. New York: Academic Press.
- For answering questions in HW6:
Dommenget and Latif, 2002:
A cautionary note on the interpretation of EOFS. J. Clim., 15,
216--225.
- Optional: Kaiser (Psychometrika 23(3),
187-200, 1958), Sherin
(Psychometrika 31(4), 535-538), Neudecker
(Psychometrika 46(3), 343-345, 1981), Jennrich
(Psychometrika 66(2), 289-306, 2001), which outline the varimax
criterion, matrix formulation, and the SVD formulation of the iterated
varimax solution.
Reading questions:
- What are the costs and benefits of rotation of a set of
leading EOFs?
- In an orthogonal rotation of a subset of EOFs such as varimax,
why are the corresponding principal components no longer orthogonal?
- How are the schematic 2-D orthogonal and 'oblique' rotations
different from the unrotated EOF representation of the data shown in
Fig. 11.11?
Products to hand in (keep a
copy for yourself to use in class discussion):
- Answers to the reading questions listed above.
Back
to Schedule/Syllabus.