R help - Feb 2005 - How to solve error : "cannot allocate vector of size 1208235 Kb"

Howdy R gurus !

I am newbie to R
I use R 2.0.1 in Windows XP. When I run R
I got the follwoing memory error.
My physical memory size is 3 Gb.
My R got the memory problem when it reached to 
about 2 Gb.

Thanks in advance,

> library(spdep)
> sfr.lagsarlm <- lagsarlm(sfr.data$Bldgsqft ~ sfr.data$Ncounty +
sfr.data$Nugb + sfr.data$Ngroup, data=sfr.data, listw=sfr.listw,
method="eigen")
Error: cannot allocate vector of size 1208235 Kb
> memory.size(max=FALSE)
[1] 17862584
> memory.limit(size=NA)
[1] 3145728000
> 

-- 
Kum-Hoe Hwang, Ph.D.

Kyonggi Research Institute, Korea (ROK)
(Urban Planning and GIS)
Phone : 82-31-250-3283
Email : phdhwang at gmail.com

You need to read the rw-FAQ (as the posting guide asks) and make the R 
executable capable of using more than 2Gb (assuming that you have already 
enabled your OS).

On Fri, 11 Feb 2005, Kum-Hoe Hwang wrote:
> Howdy R gurus !
>
> I am newbie to R
> I use R 2.0.1 in Windows XP. When I run R
> I got the follwoing memory error.
> My physical memory size is 3 Gb.
> My R got the memory problem when it reached to
> about 2 Gb.
>
> Thanks in advance,
>
>
>> library(spdep)
>> sfr.lagsarlm <- lagsarlm(sfr.data$Bldgsqft ~ sfr.data$Ncounty +
sfr.data$Nugb + sfr.data$Ngroup, data=sfr.data, listw=sfr.listw,
method="eigen")
> Error: cannot allocate vector of size 1208235 Kb
>> memory.size(max=FALSE)
> [1] 17862584
That tells you the state after that command failed: the max=TRUE figure is
more interesting at that point.
>> memory.limit(size=NA)
> [1] 3145728000
>>
>
>
> -- 
> Kum-Hoe Hwang, Ph.D.
>
> Kyonggi Research Institute, Korea (ROK)
> (Urban Planning and GIS)
> Phone : 82-31-250-3283
> Email : phdhwang at gmail.com
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

On Fri, 11 Feb 2005, Kum-Hoe Hwang wrote:
> Howdy R gurus !
> 
> I am newbie to R
> I use R 2.0.1 in Windows XP. When I run R
> I got the follwoing memory error.
> My physical memory size is 3 Gb.
> My R got the memory problem when it reached to 
> about 2 Gb.
This is not a question just about memory but also about methods. The
number of rows in sfr.data is sufficiently large (about 35000?) that the
NxN matrices being used need a lot of room. That is why the function
provides the "SparseM" method to cater for your situation (unless you
have
chosen a representation of neighbour weights that is not similar to
symmetric). 

If you choose a symmetric or similar to symmetric representation, you can
use the "SparseM" method, which, as its name suggests, uses sparse
matrices. All of this is described in help(lagsarlm). 

It is also quite possible that the last 95% of your data add almost no new
information, so you could also consider choosing a subset for analysis, in
which case the "eigen" method using dense matrices should work.
> 
> Thanks in advance,
> 
> 
> > library(spdep)
> > sfr.lagsarlm <- lagsarlm(sfr.data$Bldgsqft ~ sfr.data$Ncounty +
sfr.data$Nugb + sfr.data$Ngroup, data=sfr.data, listw=sfr.listw,
method="eigen")
> Error: cannot allocate vector of size 1208235 Kb
> > memory.size(max=FALSE)
> [1] 17862584
> > memory.limit(size=NA)
> [1] 3145728000
> > 
> 
> 
> 
-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand at nhh.no

I appreciate all who helped all.
My problem was not enough memory definietely.
First I tried smaple function that solved my problem nicely.
Second, I tried a method="sparse" option in lagsarlm model.
Comparing results led me to conclude the first trial that estimates 
better results in the lagsarlm model.

Thanks millions,

Kum

On Fri, 11 Feb 2005 13:02:53 +0100 (CET), Roger Bivand
<Roger.Bivand at nhh.no> wrote:
> On Fri, 11 Feb 2005, Kum-Hoe Hwang wrote:
> 
> > Howdy R gurus !
> >
> > I am newbie to R
> > I use R 2.0.1 in Windows XP. When I run R
> > I got the follwoing memory error.
> > My physical memory size is 3 Gb.
> > My R got the memory problem when it reached to
> > about 2 Gb.
> 
> This is not a question just about memory but also about methods. The
> number of rows in sfr.data is sufficiently large (about 35000?) that the
> NxN matrices being used need a lot of room. That is why the function
> provides the "SparseM" method to cater for your situation (unless
you have
> chosen a representation of neighbour weights that is not similar to
> symmetric).
> 
> If you choose a symmetric or similar to symmetric representation, you can
> use the "SparseM" method, which, as its name suggests, uses
sparse
> matrices. All of this is described in help(lagsarlm).
> 
> It is also quite possible that the last 95% of your data add almost no new
> information, so you could also consider choosing a subset for analysis, in
> which case the "eigen" method using dense matrices should work.
> 
> >
> > Thanks in advance,
> >
> >
> > > library(spdep)
> > > sfr.lagsarlm <- lagsarlm(sfr.data$Bldgsqft ~ sfr.data$Ncounty
+ sfr.data$Nugb + sfr.data$Ngroup, data=sfr.data, listw=sfr.listw,
method="eigen")
> > Error: cannot allocate vector of size 1208235 Kb
> > > memory.size(max=FALSE)
> > [1] 17862584
> > > memory.limit(size=NA)
> > [1] 3145728000
> > >
> >
> >
> >
> 
> --
> Roger Bivand
> Economic Geography Section, Department of Economics, Norwegian School of
> Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
> Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
> e-mail: Roger.Bivand at nhh.no
> 
> 

-- 
Kum-Hoe Hwang, Ph.D.

Kyonggi Research Institute, Korea (ROK)
(Urban Planning and GIS)
Phone : 82-31-250-3283
Email : phdhwang at gmail.com