[EM] simulations

[Juho]

On Dec 27, 2006, at 3:55 , Kevin Venzke wrote:

> Another thing (possibly of limited use) that I've implemented is the
> ability to specify a voterless hole in the center of the distribution,
> so that voters form a ring around the position being plotted. This of
> course makes FPP and IRV show their bias towards outer candidates with
> a narrower spread of voters. (That is, a candidate precisely in the
> center of all voters might not receive any first preferences.)

In addition to this there was also earlier some discussion on voter  
distribution. I just had one thought about voter distribution that I  
wanted to share with the active simulators (so far I only quickly  
compiled Brian Olson's code with error results :-) , thanks for the  
code anyway).

The thought was to start from position where the voters are  
distributed in several portions around each of the candidates so that  
the candidates are the centre points of each set of voters. The  
centre of the figure would represent this setting. Then it is  
possible to move the voters.

The interesting part (to me) was however the possibility to rotate  
the voters e.g.  by 15 degrees around the centre point. This way one  
could generate circular preferences, which would in turn make it  
possible to simulate the different Condorcet completion methods that  
have been discussed on this mailing list so much. A natural starting  
point would be to simply have three candidates in triangle around the  
centre point of the figure.

This kind of simulations may give quite complex results e.g. when  
some of the voter groupings move close to some other candidate. But  
maybe with suitable parameter settings one could demonstrate the true  
nature and differences of the Condorcet completion methods. Maybe  
others too.

Juho Laatu


		
___________________________________________________________ 
Try the all-new Yahoo! Mail. "The New Version is radically easier to use" – The Wall Street Journal 
http://uk.docs.yahoo.com/nowyoucan.html