[R] Rounding problem R vs Excel

[Marc Schwartz]

On Wed, 2003-06-11 at 01:51, Martin Maechler wrote:
> Thanks a lot, Marc, for this extra insight.
> 
> Note that Sun / Openoffice.org 's  approxEqual(a,b)
> is not even symmetric in a and b  which I think is *really wrong*
> ((S's somewhat related  all.equal.numeric() method *is* symmetric))


Glad to be of help Martin.

What would be interesting, is to see an independent review of the
statistical functions in OOo Calc and Gnumeric as has been done for
Excel. Perhaps McCullough (who has done many of the published articles
on Excel) or someone else with the time might do this at some point.

I raise this given the increasing presence of Linux and/or OOo,
especially in light of some of the recent high profile wins over MS such
as Munich. I can envision a situation where OOo Calc and Gnumeric are
increasingly used in place of Excel as the "application of choice" where
basic statistics are being taught and used.

As Cryer stated in his review of Excel:

"Friends don't let friends use Excel for statistics"

;-)


Best regards,

Marc