Dear R users,
I'm applying a correlation structure in a mixed model (nmle function) to
control for spatial correlation between land parcels that are adjacent to
each other. I generated X,Y coordinates in ArcGIS for each land parcel and
used them in the correlation form like this:
test.exp<-corExp(1, form = ~ X + Y)
test.exp<- Initialize(test.exp,dataset)
However, the correlation Matrix generated does not look right. Here is a
portion of the matrix, in which all correlations are 0:
> corMatrix(pdxspc.exp)[1:5, 1:5]
[,1] [,2] [,3] [,4] [,5]
[1,] 1 0 0 0 0
[2,] 0 1 0 0 0
[3,] 0 0 1 0 0
[4,] 0 0 0 1 0
[5,] 0 0 0 0 1
It seems I'm not using the X Y coordinates correctly in the form. I saw two
examples that used "east" and "north" in their form. But I
do not quite
know what "east" and "north" are. Any help would be greatly
appreciated.
Best
Gary
[[alternative HTML version deleted]]
