A quick idea on my way out the door. create a vector of N 1's
N#1 add a random vector of N 0's and a single 1 (i.=?)N Sum the new vector V' and exit if +/V' is equal to Q (probably a converging power conjunction) Cheers, bob On May 8, 2015, at 9:47 AM, Dan Bron <[email protected]> wrote: > I think you've got a model in your head where Q=N; in that case, yes, your > function should produce N#1 . But N and Q are independent variables. > > -Dan > > ----- Original Message --------------- > > Subject: Re: [Jprogramming] Random distribution > From: "'Pascal Jasmin' via Programming" <[email protected]> > Date: Fri, 8 May 2015 16:27:33 +0000 (UTC) > To: "[email protected]" <[email protected]> > > For Q = 3, > > only 1 1 1 would match your criteria > > for Q = 4, > 2 2 would not be N=#V > > 1 1 1 1 again appears to be the only match I see. > > > ----- Original Message ----- > From: Dan Bron <[email protected]> > To: [email protected] > Cc: > Sent: Friday, May 8, 2015 11:54 AM > Subject: [Jprogramming] Random distribution > > I have a quantity Q and I want to divide it into a vector V of N > randomly-sized chunks, such that N=#V and Q=+/V . > > Ideas? > > -Dan > > > > > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
