[R] annot have zero distances in "corSpatial"

arun lal arunkrlal at yahoo.co.in
Wed Nov 9 11:21:34 CET 2005


Hi all,

I have been attempting this repeated measure analysis
where correlation among observations (within each
subject) is spatial. 

My command is:
>fdat.lme1<-lme(Lint~Rep+Treatment,data=fdat,random=~1|fldPos,correlation=corLin(form=~lat+lon))

It is generating follwoing error:

Error in getCovariate.corSpatial(object, data = data)
: 
	Cannot have zero distances in "corSpatial"

first I thought that decimal values of
latitude/longitude are resulting in extremely small
values of distances, causing R to ignore it. So, I
scaled it up and made the coordinates integer values.
But I am still getting the same error.

Anyone can help? I don't know if you will need more
info to answer this question. Please let me know.

arun @LSU




More information about the R-help mailing list