FACTDOUBLE

Norman Harker

unread,

Jun 21, 2001, 11:36:02 PM6/21/01

to

Hi!

I'm writing some material on functions. Has anyone used the FACTDOUBLE
function?

I've looked in my maths and stats (limited) library but find no references
to it.

I know what is does (Help is quite useful). But why use it?

tia

--
Norman Harker
Senior Lecturer, Real Estate Investment Analysis and Feasibility Studies
University of Western Sydney.

Jay T. Emory

unread,

Jun 22, 2001, 9:31:03 AM6/22/01

to

Norman,
I've been an engineer for 25 yrs and have never needed a
dble factorial. Looks to be something used in discrete
variable calculus (finite differences). A brief look in
some old text books, however, didn't yield any specific
reference.

Sorry, I'm afraid this isn't much help.

Jay

>.
>

Norman Harker

unread,

Jun 22, 2001, 11:42:24 PM6/22/01

to

Thank Jay!

It seems to be very little used and I still can't find any references to
double factorials.

Sees Ya!
"Jay T. Emory" <jte...@aol.com> wrote in message
news:1cf501c0fb1f$95481120$9be62ecf@tkmsftngxa03...

Eero Tibar

unread,

Jun 23, 2001, 4:44:59 AM6/23/01

to

Factdouble is a matter of great moment:

asin x = x + sum {1*3*5*...*(2n-1)/[2*4*6*...*(2*n)*(2*n+1)]}*x^(2*n+1)

Sum from n=1 to infinity

Norman Harker wrote in message ...

Norman Harker

unread,

Jun 23, 2001, 6:00:10 AM6/23/01

to

Sorry Eero!

But are you pulling my leg?

=asin(0.5) returns 0.523598776 will do me.

Sees Ya!

"Eero Tibar" <ee...@matti.ee> wrote in message
news:9h1nk9$bgfp4$1...@ID-37884.news.dfncis.de...

Jerry W. Lewis

unread,

Jun 23, 2001, 10:33:11 AM6/23/01

to

I think his point is that it is useful in numerical analysis.

Some people whant to know how the TV works, others just want to watch
it. They don't have to know the theory to just watch TV; but the theory
still has a big impact on their life even if it is of no direct
importance to them.

Jerry

Eero Tibar

unread,

Jun 23, 2001, 12:07:06 PM6/23/01

to

Ok
If you want a real-life example then here it goes:

N couples were lined up in front of a theatre, each couple together. This
could have happened in 2*4*6*...*(2*n) ways.

Norman Harker wrote in message ...
>Sorry Eero!
>
>But are you pulling my leg?
>
>=asin(0.5) returns 0.523598776 will do me.
>
>Sees Ya!
>
>
>"Eero Tibar" <ee...@matti.ee> wrote in message
>news:9h1nk9$bgfp4$1...@ID-37884.news.dfncis.de...
>> Factdouble is a matter of great moment:
>>
>> asin x = x + sum {1*3*5*...*(2n-1)/[2*4*6*...*(2*n)*(2*n+1)]}*x^(2*n+1)
>>
>> Sum from n=1 to infinity

Norman Harker

unread,

Jun 23, 2001, 12:58:33 PM6/23/01

to

Thanks Eero!

I can really understand that application. The first I've seen and I did do a
pretty extensive web search on FACTDOUBLE.

Sees Ya!
"Eero Tibar" <ee...@matti.ee> wrote in message

news:9h2hft$bu92m$1...@ID-37884.news.dfncis.de...

David Hilberg

unread,

Jun 25, 2001, 12:45:25 PM6/25/01

to

Hmmm. According to Help, that's not the way Excel calculates
Factdouble. We agree that 3 couples can be arranged in 1*2*3*(2^3)
ways (=2*4*6), but Excel supposedly calculates Factdouble(3) as 1*3.

From Help:

Examples

FACTDOUBLE(6) equals 48

FACTDOUBLE(7) equals 105

- David Hilberg

On Sat, 23 Jun 2001 19:07:06 +0300, "Eero Tibar" <ee...@matti.ee>
wrote:

_____________
Please reply to newsgroups for follow-up; For personal messages,
remove "nöspäm" from address.

Norman Harker

unread,

Jun 25, 2001, 5:02:20 PM6/25/01

to

Thanks David!

That puts me back to square A1.

Sees Ya!
"David Hilberg" <davidh@nöspäm.ericmarder.com> wrote in message
news:3b376954...@news2.fcc.net...

David J. Braden

unread,

Jun 26, 2001, 8:09:52 PM6/26/01

to

Hm, sorry I missed this in the last days. For a reference, see the
seminal 4-volume set by Johnson and Kotz (1ed.) or the 2nd ed. by J, K
and Kemp. First volume (Univariate Discrete Distributions) has some
preliminaries for the set. Double Factorial, written as k!!, is
described on page 2.

Dave Braden

In article <zFNZ6.11642$qJ4.4...@ozemail.com.au>,

Norman Harker

unread,

Jun 26, 2001, 11:47:59 PM6/26/01

to

Thanks David!

I just knew there had to be some use for it.

"David J. Braden" <dbr...@rochester.rr.com> wrote in message
news:dbraden-841D40...@msnews.microsoft.com...

David J. Braden

unread,

Jun 27, 2001, 1:38:03 AM6/27/01

to

Er, Norman, that may be stretching it a bit. I have only seen double
factorials come up a couple of times in 20 years of stats; I wonder what
niche of problems even led to their adoption?

dave Braden

In article <2Ic_6.12375$qJ4.5...@ozemail.com.au>,

Norman Harker

unread,

Jun 27, 2001, 3:59:00 AM6/27/01

to

Hi David!

It doesn't surprise me. Apart from a fairly thorough internet search I also
asked our professors of Maths and Stats and drew a blank.

But I can't believe that the function was included without some fairly
useful application.

"David J. Braden" <dbr...@rochester.rr.com> wrote in message

news:dbraden-5CE3B2...@msnews.microsoft.com...

David J. Braden

unread,

Jun 27, 2001, 6:39:06 AM6/27/01

to

Norm,
Let's Keep in mind that FACTDOUBLE is part of the Analysis ToolPak.
Given that, I am not perplexed that something of little import appears
in my Excel setup. What a find, Norman! It's sort of like the
realization that there are so many ways to do a regression within Excel,
and that they are internally inconsistent in some cases. IMO, the
product really lost a lot of focus by adding poorly-implemented bells
and whistles instead of repairing the foundation.

Hmm, that said, I guess I can expect my MSDN shipments to get "lost"?

Dave Braden

In article <9ng_6.12986$qJ4.5...@ozemail.com.au>,