On Tue, 11 Sep 2001, Jean-Cedric Chappelier wrote:
> Does anyone knows if there is a way to handle sparse matrices in R?
Kurt Hornik and I had several discussions about this a few weeks ago.
For the moment the answer seems to be no, or rather it is up to users
to find libraries and roll their own procedures in some lower level
language. Since there are several formats for sparse matrices and
serious questions about what range of operations would be useful, it
would be nice to have some discussion about this. I've been toying
with the idea of implementing Saunders LSQR for sparse lm() problems
with the idea that it could be extended to interior point methods
for absolute error problems as well.
url: http://www.econ.uiuc.edu Roger Koenker
email roger at ysidro.econ.uiuc.edu Department of Economics
vox: 217-333-4558 University of Illinois
fax: 217-244-6678 Champaign, IL 61820
On Tue, 11 Sep 2001, Jean-Cedric Chappelier wrote:
> Does anyone knows if there is a way to handle sparse matrices in R?
>
> More precisely I am looking for a way to do correspondance analysis
> on a contingency table which is quite huge (1,000 x 10,000) but sparse
> (i.e. full of zero).
>
> Creating it directly in R (eg as x <- array(0,c(1000,10000) then filling
> the non zero cells) seems definitely not the good solution as
> R is then unable to process it (e.g. ca(x) stop for out of memory:
> "Error: cannot allocate vector of size xxx")
>
> Any help?
>
> Thank you.
> --
> Jean-Cédric Chappelier | WWW: http://liawww.epfl.ch/~chaps
> DI-LIA-INR232 | Tel: +41 21 693.66.83 or .78
> EPFL, DI (INR), CH-1015 Lausanne | Fax: +41 21 693.52.25
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
>
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._