Archiver > GEN-MEDIEVAL > 2007-08 > 1187886634

Subject: Re: Calculating The Joint Probability Of False Paternity Events [FPE]
Date: Thu, 23 Aug 2007 12:30:34 EDT

In a message dated 8/23/2007 2:46:17 A.M. Pacific Daylight Time,

OK, try this. I had vague recollections reinforced by obvious
requirements for P(n,p) at its limiting values. Obviously P(n,0.02) is
np for n=0,1 and flattens before n=50. An exact calculation fills in the
details. It would be good to see a full series of P(n,p) curves plotted
out but that seems unlikely to happen here.

Can I go now please sir? I had no idea my initial post of Aug 1 was
going to take so long.

No. Again we get this pathetic whining about why your ridiculous math should
be *passed* instead of pointed out to you. A self-proclaimed physicist,
computer scientist, mathematician or whatever pose you're taking using
*arithmetic* to replace statistical methods. Perhaps you'd like to look at my "Book
store" page where you'll be quite happy to see I have for sale a "Handbook of
Tables for Mathematics".

You have no excuse for not having one of these which would very simply and
easily tell you how probabilities progress over multiple generations. Knowing
that they do, you have no excuse for hand-waving about what *may* happen
over 25 generations and claiming that your result was "close" after-the-fact
when it's now crystal-clear that you made no attempt to do or even discuss the
appropriate math.

What you did was use your experimental, non-random results to back-create
some "mathematical model" based on fallacy, that you then claimed was a good fit
to the data. It was no such thing and for all the wrong reasons.

Your constant failure to own this error, and try to keep pretending it
didn't happen, is never going to pass. Not today, not tomorrow, not ever.

No high-school graduate, let alone scientific academic would ever make such
an enormous error. The error is not small, its enormous. It's not obscure,
it's obvious. Your logic is not straightforward, its circular. You created
a model artificially to fit your data after-the-fact and you want us to
excuse it.

You still have the chance to admit that you made a blunder and there's no
excuse for it and then move on.

Will Johnson

************************************** Get a sneak peek of the all-new AOL at

This thread: