RE: Pareto laws and expected income

From: Jonathan Colvin <jcolvin.domain.name.hidden>
Date: Tue, 21 Jun 2005 00:00:38 -0700

Russell Standish wrote:
> I retract an earlier agreement with Jonathon that the
> expected income argument is the same as the "Why I am not
> Chinese argument". They are not, for the simple expedient
> that ones income does not affect your chances of birth (if
> there is any effect, it would be a negative one with your
> parent's wealth). In the "Why am I not Chinese" argument, the
> population of your country of birth is indexical.
>
> My expected income is a complicated function of my life's
> history, it may have some bearing on things like my innate
> intelligence, and my parent's wealth, but the problem is so
> multifactorial it is inappropriate to use anthropic reasoning.

My consciousness (or degree of such) is a complicated function of my
evolutionary history, but the problem is so multifactorial it is
inappropriate to use anthropic reasoning.

>
> What it does show is what an ass the ASSA is. It is
> unreasonable to suppose that my current wealth is sampled
> randomly from the distribution of al wealths (a Pareto
> distribution like P(x)=x^a, for some a).

Why is it any more unreasonable than supposing that your birth rank is
sampled randomly from the distribution of all birth ranks?

>
> A more interesting point that Jonathon Colvin could have made
> was questioning why ones IQ is so high. It is a reasonable
> speculation that the IQ of people on this list would usually
> be far above average (IQ=100 by definition). Of course, that
> is selective effects of this list. But one can also ask why
> in anthropic reasoning is my IQ in the top part of the
> distribution (I don't know my IQ, but I'm sure I'm in the
> tail :). The answer is that if you consider all possible
> congenital characteristics (eg country of birth, parents
> wealth, intelligence, skill at playing ball, etc.), there is
> very likely one or two chracteristics that are extreme. In my
> case it happens to be intelligence. My family's income was
> below average (for Australia that is, but probably more on a
> par with world average, actually)

If you consider all possible personal characteristics, there is very likely
one or two characteristics that are extreme. In my case it happens to be
birth rank.

>
> So that is not so strange really. However, when it comes to
> sampling indexical quantities (eg birth rates or population
> sizes), anthropic arguments take on a particular force, than
> sampling non-indexical quantities.

Why do anthropic arguments suddenly take on such force when sampling
indexical quantities?

Jonathan Colvin
Received on Tue Jun 21 2005 - 03:05:27 PDT