[EM] Why is Bayesian Regret the "Gold Standard"

Howard Swerdfeger electorama.com at howard.swerdfeger.com
Thu Mar 15 06:04:02 PDT 2007

> What is the justification for Bayesian Regret, as used in IEVS and
> described at http://rangevoting.org/BayRegDum.html, being the
> "uniquely right" metric, the "gold standard", for comparing different
> election methods or varying election scenarios?

I think this is a good question and I don't exactly want this thread to 
die. but I don't know the answer. I think whoever wrote that thinks it 
is the gold standard, other people might Look to Arrow.

> Why is the societal utility for a candidate the sum of the voters'
> individual utilities for that candidate?  How does that avoid the
> problems of being arbitrary or not well-defined, when making
> interpersonal comparisons or summations of von Neumann-Morgenstern
> utilities?  What is Bayesian about the resulting value, which is just
> a difference in utility values?

I am personally new to this measure, and I haven't read the paper they 
refer to (soon hopefully). but I wonder how they generated the personal 
Utilities for Voter V and Candidate X. you can place them both in Issue 
space, and measure there distances R. but then you need to relate 
Personal Utility with distance.
  U(V, X) = R(v, x)
or perhaps is better. as you might really Hate people who are really 
  U(V, X) = Exp ( R(v, x) )

and then you have the Social Utility SU, they do say that
  * SU = Sum(U(V, X))
But I think something Like
  * SU = Sum( Log (U(V, X)) )

Would be defendable, as we might want to integrate the extreme points of 
view into our society. and try to make them happy.

Anyway I think a good discussion about voter regret and how you measure 
it is very appropriate.

> Matthew Welland matt at kiatoa.com on Sun Mar 11 09:44:32 PDT 2007,
> wrote: 
> (http://lists.electorama.com/pipermail/election-methods-electorama.com/2007-March/019827.html)
>> ... I know I should be using  Bayesian Regret but a)  don't really
>> understand it and ...
> Warren Smith wds at math.temple.edu  on Fri Feb 9 13:06:01 PST 2007,
> wrote:
> (http://lists.electorama.com/pipermail/election-methods-electorama.com/2007-February/019439.html)
>> I happen to think Bayesian regret is a good metric, in fact the
>> uniquely right metric, not "the wrong thing". 

More information about the Election-Methods mailing list