A Google on this reveals implementations in R
and also a hint that some flavour of -robvar-
might be helpful. At this moment I have no access
to (manuals on) Stata 6, but -robvar- was added to
Stata about then.
As nothing of ancient vintage is evident, you face
a further problem in that many programmers would find
it difficult to write for Stata 6 even if they got
interested in your problem: there are so many features
that would have to be avoided.
More positively, a recipe for testing differences in
scale is this:
1. Compute absolute deviations =
| values - favourite group summary (e.g. median)|.
2. Apply favourite test of location to absolute deviations.
Nick
[email protected]
Rupert Ward
Does anyone on this list know of an implementation in
stata of Fligner-Killeen's nonparametric test for differences
in variation? (see refs below)
I have Stata 6 and have done the obvious searches to
no avail. It occured to me that someone else on this
list might have been through the same thing and be able
to assist.
refs
1. Steven M. Donnelly and Andrew Kramer. 1999. "Testing
for multiple species in fossil samples: An evaluation
and comparison of tests for equal relative variation."
American Journal of Physical Anthropology. 108: 507-529.
2. Fligner and Killeen ([1976] J. Am. Stat. Assoc. 71:210-213
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
