[EM] simulations

[Kevin Venzke]

Hi,

I posted a few new pictures
http://www.geocities.com/stepjak/index.htm

In particular, I wanted to implement approval-with-poll but with a less
useful poll. Rather than learning the top two candidates, the voters
learn which candidates receive under 1/3 of the winning score. They then
disregard these candidates, find the new mean without them, and approve
above it.

As an estimate of how much information real voters would have, I think
this is stingy. But the result already is not that bad, in my opinion.

I also corrected my idea of a cloneproof zero-info strategy. Voters
approve candidates better than the average of the one best and one worst
candidate.

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.)

Kevin Venzke


	

	
		
___________________________________________________________________________ 
Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions ! 
Profitez des connaissances, des opinions et des expériences des internautes sur Yahoo! Questions/Réponses 
http://fr.answers.yahoo.com