[R-sig-Geo] Spatial correlation for nominal variable

[Marco Helbich]

Dear Monir,

perhaps the join count statistic for k coloured factors in the spdep package is a possibility.

best regards
marco
 
> Message: 2
> Date: Mon, 31 Aug 2009 08:26:43 -0500
> From: "Hossain, Md Monir" <Monir.Hossain at uth.tmc.edu>
> Subject: [R-sig-Geo] Spatial correlation for nominal variable
> To: "r-sig-geo at stat.math.ethz.ch" <r-sig-geo at stat.math.ethz.ch>
> Message-ID:
> 	<FE012FDB2E167148A07CBDE554B3F3A1093DCF0AB3 at UTHCMS2.uthouston.edu>
> Content-Type: text/plain
> 
> Dear all,
> I would appreciate much for sharing your thoughts in regard to spatial
> correlation for nominal variable.
> I have a variable named FP_Pattern for FringerPrint Pattern Designation
> for IS6110 to indicate genotype. This is a nominal variable and has 17
> levels. For each TB patient, I also have longitude and latitude. The purpose of
> analyzing this data is to answer the question whether for given level of
> genotype is there any spatial correlation.
> 
> Thanks in advance for your time,
> 
> Monir
> 

-- 
Jetzt kostenlos herunterladen: Internet Explorer 8 und Mozilla Firefox 3 -
sicherer, schneller und einfacher! http://portal.gmx.net/de/go/atbrowser