Date:         Tue, 25 Jul 2000 13:42:47 -0700
Reply-To:     "William W. Viergever" <wwvierg@IBM.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "William W. Viergever" <wwvierg@IBM.NET>
Subject:      Re: 10 random numbers
In-Reply-To:  <00365BB3.C21292@westat.com>
Content-Type: text/plain; charset="us-ascii"

At 03:54 PM 07/25/2000, HERMANS1 wrote:
>It seems to me that division of each of a sequence of random random numbers by
>their sum amounts more or less to normalizing or rescaling the sequence.  The
>afine scaling factor, in this case the reciprocal of the sum of the sequence
>n[i], depresses the values of each element in the sequence so that if drawn from
>a uniform distribution n[2] lies in the interval (0,1-n[1]), n[3]  in
>(0,1-(n[1]+n[2])) ....., with no degrees of freedom left for n[10].

Bet you can't say that 10 times real fast! <g>
----------------------------------------------------------------------------
William W. Viergever                      Voice  : (916) 483-8398
Viergever & Associates                    Fax    : (916) 486-1488
A SAS Institute Quality Partner (USA)     E-mail : wwvierg@ibm.net
Sacramento, CA 95825

"When you have excluded the impossible, whatever remains, however improbable, must be the truth."
- S. Holmes

"Act with courage and dignity; stick to the ideals that give meaning to life."
- J. Nehru
----------------------------------------------------------------------------