[R-sig-Geo] EBlocal - "variable contains non-finite values"

[Roger Bivand]

On Thu, 17 Mar 2011, Mathieu Rajerison wrote:

> Hi list,
>
>
> I performed the EBlocal function on my dataset, but I got the following
> error:
>
>> res <- EBlocal(agg$NPARBAT, agg$Expected, nb)
> Erreur dans lag.listw(lw, (ni * (xi - m.i)^2), zero.policy = zero.policy) :
>  Variable contains non-finite values
>

Almost certainly at least one of the values in agg$Expected is zero. 
Indeed, this should not be the expected count, but rather the populations 
at risk, as explained in ?EBlocal, which should all be positive numbers. 
So you are most likely not using the function correctly.

Hope this helps,

Roger

>
> I noticed someone else got the same error message as in this message:
> http://www.mail-archive.com/r-sig-geo@stat.math.ethz.ch/msg07922.html
>
>
> Apparently, it was fixed in the latest version of spdep. My version is
> 0.5-29 (the latest, I think) as the sessionInfo() indicates:
>
>> sessionInfo()
> R version 2.11.1 (2010-05-31)
> i386-pc-mingw32
>
> (...)
>
> other attached packages:
> [1] rgdal_0.6-28       spdep_0.5-29       coda_0.14-2
> deldir_0.0-13
> [5] maptools_0.7-38    foreign_0.8-40     nlme_3.1-96
> MASS_7.3-6
> [9] Matrix_0.999375-39 lattice_0.18-8     boot_1.2-42
> sp_0.9-76
>
> loaded via a namespace (and not attached):
> [1] grid_2.11.1  tools_2.11.1
>
>
> Any help would be greatly appreciated!
>
> Mathieu
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> 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.Bivand at nhh.no