Post Reply 
F distribution on the 41C
12-26-2015, 03:38 PM
Post: #41
RE: F distribution on the 41C
(12-20-2015 06:59 AM)Ángel Martin Wrote:  
(12-20-2015 12:28 AM)quantalume Wrote:  Do you know anything about Ángel's XMSTAT ROM? All I can find is a quick-reference guide.

The XMSTAT module is in flux at the moment. The QRG shows that it contains all those distributions (Chi2, Normal, T-Student, F-Snedecor, and Binomial) - and lots of other statistical functions.
'AM


Gene: Hey Angel. The XMSTAT module contains the LR and LRY functions (thank you very much), but how about an LRX function? That would use the LR relationship to compute X given Y (the reverse of LRY). I'm not aware of this function existing anywhere else, and it is often useful in statistical environments. "In what month will sales equal $Y?"

Looked like the FAT table had an open entry and the code can't be that big. :-)
Find all posts by this user
Quote this message in a reply
12-26-2015, 04:56 PM (This post was last modified: 12-26-2015 04:57 PM by Dieter.)
Post: #42
RE: F distribution on the 41C
(12-26-2015 03:38 PM)Gene Wrote:  Gene: Hey Angel. The XMSTAT module contains the LR and LRY functions (thank you very much), but how about an LRX function? That would use the LR relationship to compute X given Y (the reverse of LRY). I'm not aware of this function existing anywhere else, and it is often useful in statistical environments. "In what month will sales equal $Y?"

Hmm... I wonder why you're not aware of such a function on other calculators: Both the 35s and the 34s feature this function (y-hat as well as x-hat). So does the 30b (Pred X and Pred Y). Even TI did it in the Seventies on the 58/59 (OP 14 and OP 15). And I assume these are not the only calculators that do it both ways.

Dieter
Find all posts by this user
Quote this message in a reply
12-26-2015, 05:19 PM
Post: #43
RE: F distribution on the 41C
(12-26-2015 03:03 PM)Namir Wrote:  Thanks for the update Dieter.

For a much more sophisticated version you may want to take a look at the alglib library. In some versions (I got 2.6.0 here) the source code is available in Visual Basic (here VB6). The "src" directory contains lots of code for special functions, among them ibetaf.bas and betaf.bas.

Dieter
Find all posts by this user
Quote this message in a reply
12-26-2015, 06:04 PM
Post: #44
RE: F distribution on the 41C
(12-26-2015 04:56 PM)Dieter Wrote:  
(12-26-2015 03:38 PM)Gene Wrote:  Gene: I'm not aware of this function existing anywhere else...
Hmm... I wonder why you're not aware of such a function on other calculators:
Dieter

Gene: My fault, Dieter. I was sloppy in my response. I meant to say this:

"I'm not aware of this function existing anywhere else in any of Angel's HP 41CL roms dealing with statistics or math."

:-)

I am aware of its existence on other machines. Sorry for my sloppy writing.
Find all posts by this user
Quote this message in a reply
12-26-2015, 10:09 PM
Post: #45
RE: F distribution on the 41C
(12-22-2015 09:34 PM)Paul Dale Wrote:  [ Implementing the regularized Beta function ]
I don't think I'd seen the DiDonato and Jarnagin reference before.

A PDF of the paper is freely available. Maybe you can also take a look at a "classic" implementation: Algorithm AS 63 by Majumder and Bhattacharjee, published 1973 in "Applied Statistics". The original Fortran code can be found in various places, at the moment I am doing some tests in VBA with a slighty modified version. This code seems to require more iterations, but each does just a few operations.

However, IMHO the continued fraction approach posted earlier in this thred seems to work just as well. In both methods the crucial point seems to be the threshold where the switch between the calculation of Ix(a,b) and I1-x(b,a) occurs.

Dieter
Find all posts by this user
Quote this message in a reply
12-30-2015, 08:44 AM
Post: #46
RE: F distribution on the 41C
(12-26-2015 03:38 PM)Gene Wrote:  Hey Angel. The XMSTAT module contains the LR and LRY functions (thank you very much), but how about an LRX function? That would use the LR relationship to compute X given Y (the reverse of LRY). I'm not aware of this function existing anywhere else, and it is often useful in statistical environments. "In what month will sales equal $Y?"

Looked like the FAT table had an open entry and the code can't be that big. :-)

Done, look out for the updated version coming soon.

Cheers,
'AM

"To live or die by your own sword one must first learn to wield it aptly."
Find all posts by this user
Quote this message in a reply
12-30-2015, 08:46 AM
Post: #47
RE: F distribution on the 41C
(12-26-2015 06:04 PM)Gene Wrote:  
(12-26-2015 04:56 PM)Dieter Wrote:  Hmm... I wonder why you're not aware of such a function on other calculators:
Dieter

Gene: My fault, Dieter. I was sloppy in my response. I meant to say this:

"I'm not aware of this function existing anywhere else in any of Angel's HP 41CL roms dealing with statistics or math."

:-)

I am aware of its existence on other machines. Sorry for my sloppy writing.

Now also available on the 41 platform. It was very easy to add this entry on top of the LR code.

'AM

"To live or die by your own sword one must first learn to wield it aptly."
Find all posts by this user
Quote this message in a reply
12-30-2015, 05:54 PM
Post: #48
RE: F distribution on the 41C
Yay! Bravo! I hoped it would be.
Find all posts by this user
Quote this message in a reply
Post Reply 




User(s) browsing this thread: 4 Guest(s)