John: if by nested you mean hierarchical and if what you are working
with (some function of what originally were counts) may ostensibly be
viewed as normal, then you should be able to do this in SAS' PROC MIXED.
if your data remain counts then you may be able to do the same in the SAS
macro
|
| ||
|-+
--|
|
|
| To: Brian R Gray [EMAIL PROTECTED
you could modify the suggested approach by using a generalization of the
Poisson, the neg binomial assumption you mention. most stat software
allows negative binomial regression. in this case, the variance component
of the Chi-squared resids may be better approximated (than under the
Poisson
you might try a tobit assumption. brian
Brian Gray
USGS Upper Midwest Environmental Sciences Center
2630 Fanta Reed Road, La Crosse, WI 54602
608-783-7550 ext 19 - Onalaska campus or
608-781-6234 - La Crosse campus
fax
I am curious about the use of 100 realizations to generate a probability
map. is this a standard approach? if so, is a small p-value (such as
.05) used? if so, it would seem like 100 iterations might be a smallish
sample size for distinguishing, say, .05 (ie 5 outcomes out of 100) from,
say,
Hi. I am ecologist, too. While I am not a geostatistician, I don't recall
seeing strong arguments to suggest that some processes lead to one spatial
model rather than others. Of course, one might argue the other way
around--that some models might indicate more or less covariance at, say,
are you working with nominal or ordinal data? if you are interested in
inferences on the mean and if spatial correlation isn't exceptional, then
you can lump correlation with other sources of overdispersion and use an
overdispersion estimate to adjust the expected variance (ie
psi*pi(1-pi)/n).
The Biological Resources Division of the USGS is looking for a
stat-oriented ecologist to work on modeling songbird counts and
presences/absences as functions of habitat measures. Experience with
modeling spatially and/or temporally correlated count or binary data,
wildlife data and/or with