This site was mentioned in an earlier post.  If you type SAS as the search
string there are over 30 statistical papers there and most appear to include
the code: http://www.jstatsoft.org/search

Art

On Jan 11, 3:12=A0pm, art...@NETSCAPE.NET (Arthur Tabachneck) wrote:
> This site was mentioned in an earlier post. =A0If you type SAS as the sea=
rch
> string there are over 30 statistical papers there and most appear to incl=
ude
> the code:http://www.jstatsoft.org/search
>
> Art

Great site Art, thanks

 I think we need to integrate R into SAS proper, not just IML. It
looks like more than 75% of the code is R or R packages!

On Jan 11, 6:31=A0pm, xlr82sas <xlr82...@aol.com> wrote:
> On Jan 11, 3:12=A0pm, art...@NETSCAPE.NET (Arthur Tabachneck) wrote:
>
> > This site was mentioned in an earlier post. =A0If you type SAS as the s=
earch
> > string there are over 30 statistical papers there and most appear to in=
clude
> > the code:http://www.jstatsoft.org/search
>
> > Art
>
> Great site Art, thanks
>
> =A0I think we need to integrate R into SAS proper, not just IML. It
> looks like more than 75% of the code is R or R packages!

What does integration mean? If you could use SAS to create your
datasets then have R pick them up and run, does that suffice?
Likewise, if R could write a SAS dataset which you could then use in
SAS, would that be useful?

In reality, R cannot create a SAS dataset or read one in so that is
the basis of my question.

Alan
http://www.savian.net

Alan,

I can't speak for Roger, and my original post wasn't about R, but R
can do a number of things that haven't as yet been incorporated into
SAS.  Why shouldn't there be more support for doing such
capabilities?  Last time I checked, Excel doesn't offer a way to read
SAS files, but SAS still supports Excel and even offers a way to meet
such a need.

I think the entire analytical community would have appreciated such
integration of SAS and R.

Art
----------
On Jan 11, 8:57 pm, Savian <savian....@gmail.com> wrote:
> On Jan 11, 6:31 pm, xlr82sas <xlr82...@aol.com> wrote:
>
> > On Jan 11, 3:12 pm, art...@NETSCAPE.NET (Arthur Tabachneck) wrote:
>
> > > This site was mentioned in an earlier post.  If you type SAS as the search
> > > string there are over 30 statistical papers there and most appear to include
> > > the code:http://www.jstatsoft.org/search
>
> > > Art
>
> > Great site Art, thanks
>
> >  I think we need to integrate R into SAS proper, not just IML. It
> > looks like more than 75% of the code is R or R packages!
>
> What does integration mean? If you could use SAS to create your
> datasets then have R pick them up and run, does that suffice?
> Likewise, if R could write a SAS dataset which you could then use in
> SAS, would that be useful?
>
> In reality, R cannot create a SAS dataset or read one in so that is
> the basis of my question.
>
> Alanhttp://www.savian.net