On Thu, 18 Oct 2007, Dvora-Laiô Wulfsohn wrote:
Hi
I am (newly) testing out the package spgwr to run GWR in R v. 2.5.1.
I am trying to run the gwr() function with the following command:
model1 <- gwr(Nfix.ma~peaDM.ma, data=gwr.dat, coords=gwr.loc, bandwidth=200,
+ gweight=gwr.bisquare, hatmatrix=TRUE, se.fit=TRUE)
and get no output, but only the Error message:
Error in SpatialPointsDataFrame(coords = fit.points, data = df, proj4string =
CRS(p4s)) :
row.names of data and coords do not match
However, a check of rownames() in both my data (gwr.dat) and coords (gwr.loc)
reveals no differences between the two.
Perhaps, the error is due to something else altogether related to coordinates
in the sp package?
My gwr.loc is a 55x2 matrix (Easting, Northing, 55 locations) and the gwr.dat
is a 33x55 data.frame (with the first two columns also containing the Easting,
Northing data).
Please try to coerce gwr.dat to a SpatialPointsDataFrame, and omit the
coords= argument:
coordinates(gwr.dat) <- c(1,2)
should do it. The coords= argument is only retained for backward
compatibility. If you like, you can send the file from:
save(gwr.dat, gwr.loc, file="rownames_problem.RData")
to me off-list, so that I can trap this particular problem.
Roger
Can anyone advise?
Thanks,
Dvor
[[alternative HTML version deleted]]
_______________________________________________
R-sig-Geo mailing list
[email protected]
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: [EMAIL PROTECTED]
_______________________________________________
R-sig-Geo mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
