AMBER Archive (2004)

Subject: Re: AMBER: Continued:No skew or curtosis when zero variance in moment

From: Elijah Gregory (ejgregory_at_gmail.com)
Date: Tue Oct 26 2004 - 08:48:55 CDT

If you have zero variance (i.e. zero second statistical moment) then
your standard deviation is zero and calculating the skew (third
statistical moment) and curtosis (fourth statistical moment) involves
a "divide by zero" which is undefined.

The output message is to let you know that the skew and curtosis of
the probability distribution are undefined when the variance equals
zero.

But I'm just guessing here, one of the more experienced users or
developers should have a more concerete answer.

~Elijah Gregory
  Center for High Performance Computing
  University of Utah

On Tue, 26 Oct 2004 13:2:31 +0800, Xiao He <hx_at_itcc.nju.edu.cn> wrote:
> Dear Amber users,
>
> I'm terribly Sorry.It happened in NMODE program
> No skew or curtosis when zero variance in moment
>
> It is in mm_pbsa/mm_pbsa_statistics.pm line 1785,but I do not know what does this mean,thanks!
>
> Xiao He
> hx_at_itcc.nju.edu.cn
> 2004-10-26
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu