[R] large numbers of observations using ME() of spdep

Emmanuel Charpentier charpent at bacbuc.dyndns.org
Sun Jun 7 00:03:56 CEST 2009


Le dimanche 07 juin 2009 à 02:50 +0900, lucero mariani a écrit :
> Dear All,
> We aim to remove the spatial structure of our data using Moran Eigen
> Vectors and spdep package . Our data has 3694 samples and 13
> variables.
> The computer stop working after almost 4 days of processing (we found
> it emitting a sharp sound and with all colors on the screen. No
> wories, it was restared without problem!).

Cool ! Farm it out to a night club (and get a real computer with the
proceedings).

Sorry. Couldn't resist !

And twice sorry to be unable to help with your *real* problem...

					Emmanuel Charpentier

>                                             And we are left with
> nothing: no result file was produced since the calculations were
> interrumpted!
> Consequently, I am looking for a way to accelerate calculations with
> ME() of spdep and to save the results step by step (for each eigen
> value).
> I came accross several promissing ideas: 1)make a compiled program in
> C and call it in R with the command system(); 2)use muliR. However, I
> do not know how to use these tools. I do not understand exactely why
> the calculations are so slow and I do know know how to write a program
> in C...
> I am using Ubuntu 9.04 and R version 2.8.1 .
> 
> The code in R is a fallows:
> taimin.xy = taimin[c("dy", "dx")];
> coordinates(taimin.xy) <- ~dy+dx; #dy, dx, name of the respective
> coordinate columns
> coords<-coordinates(taimin.xy);
> library(spdep);
> TaiminGabrielGraph<-gabrielneigh(coords, nnmult = 12);
> tai.gab.nb<-graph2nb(TaiminGabrielGraph,sym=TRUE);
> nbtaim_distsg <- nbdists(tai.gab.nb, coords);
> nbtaim_simsg <- lapply(nbtaim_distsg, function(x) (1-((x/(4*50))^2)) );
> MEtaig.listw <- nb2listw(tai.gab.nb, glist=nbtaim_simsg, style="B");
> sevmtaig <- ME(Presabs
> ~Age+Curv+Zon2005+ZoneMeiji+Wi+Sol+Slope+Seadist+Elev+Basin,data=taimin,
> family=binomial,listw=MEtaig.listw)
> 
> Any help is welcome!
> 
> Thanks
> 
> Lucero Mariani, Yokohama National University, Japan.
> 




More information about the R-help mailing list