I haven't looked at this yet. But it sounds great. I'll try and look at it soon, but I've been a bit busy recently (actually, I might need to do some regression analysis shortly).
In any event, I haven't got any problems with Jason having CVS access.
J'
On Tue, Aug 16, 2005 at 01:05:20AM +0000, Jason Stover wrote:
www.sakla.net/linreg.tar.gz
www.sakla.net/pspp_contrib.tar.gz
The first has a linear regression library, the second has
regression.q and some routines to recode categorical variables
to vectors. They aren't finished, but they do compile and run
correctly when called within regression.q. See regress_categorical.stat
for sample pspp syntax.
If this is the kind of thing you want, then now may be the
time for CVS access on savannah. It's getting big enough that
updating my crusty old anonymous checkout breaks everything and
takes a lot of fixing.
I'm not happy with the categorical recode routines, and I don't
think I ever will be. I suspect a less offensive way to implement
them would be to put the struct recoded_categorical inside a struct
variable, but this wouldn't eliminate the need for some of the
ugly code in cat.c.
-Jason
--
[EMAIL PROTECTED]
SDF Public Access UNIX System - http://sdf.lonestar.org
_______________________________________________
pspp-dev mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/pspp-dev
--
PGP Public key ID: 1024D/2DE827B3
fingerprint = 8797 A26D 0854 2EAB 0285 A290 8A67 719C 2DE8 27B3
See http://pgp.mit.edu or any PGP keyserver for public key.
signature.asc
Description: Digital signature
_______________________________________________ pspp-dev mailing list [email protected] http://lists.gnu.org/mailman/listinfo/pspp-dev
