[EM] More Ossipoffness...

Michael Ossipoff mikeo2106 at msn.com
Thu Jan 18 06:19:42 PST 2007


Warren said:

Second, while I happen to think Mike Ossipoff has some talents,
and indeed in the present case it turns out that his "bias free method" can
indeed be derived and explained
in a semi-reasonable way

I reply:

Warren hasn't supported his claim that Bias-Free has only been derived and 
explained in a semi-reasonable way.

Warren continues:

(albeit, at first I thought it was arrant nonsense)
his ability to explain mathematics clearly, often falls short.
If you have an algorithm to present, then by all means do so. Step 1, step 
2, etc.
Do not leave the algorithm to the reader to divine


I reply:

When you state that a method is a divisor method, and tell how its roundoff 
point is determined, then you have completely defined the method. I have 
defined Bias-Free in that way.

I have posted detailed instructions for Cycle-Webster and Adjusted-Rounding.

In thie following paragraph, to avoid overusing the words "brief and 
concise", let it be undestood that, in that paragraph, whenever I speak of 
describing, I am refering to describing briefly and concisely.

Warren has never described his method's definition in posting here. He has 
never described the goal of the derivation of his method here. He has never 
described, in outline, the derivation of his method here. When someone is 
unable to say what he is doing, we're forced to conclude that he doesnt't 
know what he is doing.


Warren continues:

Third, in his recent series Mike has made some errors, for example asserting 
the wonderfulness
of the probability density  1/(Ax+B)

I reply:

First, Warren, in typical sloppy error-prone fashion, has miscopied the 
expression.
It's B/(x+A). Actually, as I use it, it's B/(q+A).

Secondly, I didn't say that it was wondrous. I merely said that it would do. 
It will have to do, for weighting Bias-Free, because an exponential weight 
results in an integrand whose antiderivative doesn't have an exact solution.

I'd better mail this before this computer loses the Internet again. To be 
continued.

Mike Ossipoff





More information about the Election-Methods mailing list