R devel - Apr 2003 - pnorm conditional (PR#2883)

--=-YFjXKq8/D/t1qWmIzQ9D
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

I was going over the source in src/nmath/pnorm.c and noticed a little
bug in pnorm_both (in R 1.7.0).  The else-if on line 205 covers the
entire real line.  Seems you want an &&, not an ||.  Doesn't make a
big
difference (you still get a 0 or 1 from extreme starting values) but
your log refinements will always be skipped.

--=20
Daniel Pemstein
Email: dbpemste@artsci.wustl.edu, dan@lcws.org
Web: http://www.danpemstein.com
PGP: http://www.danpemstein.com/files/pgp.txt

--=-YFjXKq8/D/t1qWmIzQ9D
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iD8DBQA+sAX6ItxEqhMqILgRAjKxAJ40B1Laa6AEiYt+Mh29FhT3o03IsACfXwKb
/so6cYytUi1PmuPwZOMs75E=QgyI
-----END PGP SIGNATURE-----

--=-YFjXKq8/D/t1qWmIzQ9D--

>>>>> "Daniel" == Daniel Pemstein
<dbpemste@artsci.wustl.edu>
>>>>>     on Wed, 30 Apr 2003 19:20:36 +0200 (MET DST) writes:
    Daniel> I was going over the source in src/nmath/pnorm.c and
    Daniel> noticed a little bug in pnorm_both (in R 1.7.0).
    Daniel> The else-if on line 205 covers the entire real line.
    Daniel> Seems you want an &&, not an ||.  Doesn't make a big
    Daniel> difference (you still get a 0 or 1 from extreme
    Daniel> starting values) but your log refinements will
    Daniel> always be skipped.

Thank you, Daniel!
You are right about the || typo.  However as I found things are
a bit more subtle.  The last clause now always has been
skipped and the if( || ) was James Rath's contribution
subsequent to his PR#699 , including the mentioned asymptotic
expansion for the log_p clause.   
Now because of the typo that log_p case had not ever been in use
in the past and I know found that part the "*ccum = -0." clearly
is sub-optimal and worse than the current log_p behavior.

Anyway, this is being fixed for 1.7.1, but the patch will be
more than the 2-letter replacement,  s/||/&&/

Please let me know (off the mailing list) if you're interested
in more details.

Thanks again.
Martin Maechler <maechler@stat.math.ethz.ch>
http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO C16	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><