On Mon, 21 Jun 2010, elaine kuo wrote:
Dear ,
This is Elaine.
I am computing moran's I using moran.test for
a generalized linear model (multiregression).
Don't do this, the expected value and the calculated variance under
normality and randomisation are wrong. Use lm.morantest(), but be aware
that it is for an lm() fit, not a glm() fit. There are no tests with
statistical authority for glm residuals (as far as I know).
One would usually assign the output of a test to an object, then examine
the object with str() to find the required value, here the list component
called estimate, first vector element.
So lm.morantest(modg1, modg1.listw)$estimate[1] is the calculated value of
Moran's I for your glm model, for what it is worth.
Roger
The following contents are the results, and I cannot find the observed
Moran's I mentioned as estimate in the manual.
Please kindly help indicate if there is observed Moran's I did not notice or
other method for calculation.
Thanks
Elaine
data: residuals(modg1)
weights: modg1.listw
Moran I statistic standard deviate = 786.6486, p-value < 2.2e-16
alternative hypothesis: greater
sample estimates:
Moran I statistic Expectation Variance
6.351413e-01 -2.052545e-04 6.523188e-07
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
--
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: roger.biv...@nhh.no
_______________________________________________
R-sig-Geo mailing list
R-sig-Geo@stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo