[R-sig-Geo] : Moran I and General Trend

[Nicholas Lewin-Koh]

Hi,
Non-constant mean is a very different beast from nonconstant variance
(heteroscedasticity).
Non-constant mean can be fixed by detrending and working on the
residuals,
and thinking not to deeply about it I would imagine that as long as the 
mean wasn't to "jumpy" across the region global Moran would preform
perfectly 
well for decribing how correlated neighboring rates are.

Nicholas
On Wed, 24 Aug 2005 21:36:39 +0100 (BST), "Sotos" <sotosko at yahoo.gr>
said:
> Many thanks for you reply, 
> 
> But wouldn't Local moran be better for checking for
> autocorellation and local clusters?
> 
> Also I read that in SIDS dataset, Freeman-Tuckey
> square root transformation was used because the
> variance was not constant. Would this be appropriate
> to use for crime rates with non constant mean? I guess
> that a lot depends on the actual data, but are methods
> appropriate or am I missing something?
> 
> Cheers,
> S.
> 
> 
> 
> --- Nicholas Lewin-Koh <nikko at hailmail.net> &#941;&#947;&#961;&#945;&#968;&#949;:
> 
> > Hi
> > If you have a first order effect like that then you
> > can 
> > try fitting a linear model to the data and the use
> > the residuals for checking autocorrelation. This
> > will
> > introduce some bias but it is small relative to the
> > gain if
> > the residuals are stationary. 
> > 
> > Nicholas
> > 
> 
> 
> 
> 
> 	
> 
> 	
> 		
> ___________________________________________________________ 
> &#935;&#961;&#951;&#963;&#953;&#956;&#959;&#960;&#959;&#953;&#949;&#943;&#964;&#949; Yahoo!; 
> &#914;&#945;&#961;&#949;&#952;&#942;&#954;&#945;&#964;&#949; &#964;&#945; &#949;&#957;&#959;&#967;&#955;&#951;&#964;&#953;&#954;&#940; &#956;&#951;&#957;&#973;&#956;&#945;&#964;&#945; (spam); &#932;&#959; Yahoo! Mail 
> &#948;&#953;&#945;&#952;&#941;&#964;&#949;&#953; &#964;&#951;&#957; &#954;&#945;&#955;&#973;&#964;&#949;&#961;&#951; &#948;&#965;&#957;&#945;&#964;&#942; &#960;&#961;&#959;&#963;&#964;&#945;&#963;&#943;&#945; &#954;&#945;&#964;&#940; &#964;&#969;&#957; &#949;&#957;&#959;&#967;&#955;&#951;&#964;&#953;&#954;&#974;&#957; 
> &#956;&#951;&#957;&#965;&#956;&#940;&#964;&#969;&#957; http://login.yahoo.com/config/mail?.intl=gr 
>