Re: [office-formula] Fwd: Why is TDIST a "right tail" type?

["Andreas J. Guelzow" <andreas.guelzow@concordia.ab.ca>]

On Thu, 2010-04-08 at 05:35 -0600, Eike Rathke wrote:
> Hi,
> 
> This came on the comment list, any thoughts?

I believe the commenter is absolutely right. TDIST should become
LEGFACY.TDIST and a new left-tail TDIST should be added.

Note that Excel 2010 introduced T.DIST which is such a left-tail TDIST.
Gnumeric has had for a long time R.PT which is a left tailed CDF for the
t distribution. (In case you wonder about the unusual name, this is
derived from the naming scheme used by the R-project.) 

Andreas
> 

>   Eike
> 
> ----- Forwarded message from Regina Henschel <rb.henschel@t-online.de> -----
> 
> Date: Fri, 02 Apr 2010 22:35:58 +0200
> Subject: [office-comment] {office-formula]Why is TDIST a "right tail" type?
> To: office-comment@lists.oasis-open.org
> Message-id: <4BB6552E.9040105@t-online.de>
> 
> Hi members,
> 
> in the definition of TDIST the missing integral has been added recently.  
> But I'm wondering, that it is a "right tail" type. All other cumulative  
> distribution functions are of "left tail" type. CHIDIST and FDIST, which  
> are right-tailed in some application, where made LEGACY to get consistent 
> definitions. I have expected, that the same would be done with TDIST.
> 
> kind regards
> Regina
> 
> ----- End forwarded message -----
>