William Dickens wrote:
> >>>Well what I suppose we should be using isn't either the SD or the Var, but the %
>of the maximum increase in utility that is possible with increasing diversification.
>Playing around with a few examples it looked to me that the gain in utility was
>inversely proportional to the decline in variance - - not the SD. However, more
>surprising than that was the incredibly small utility gain that one obtains by
>reducing the standard deviation of your income from say 20% of the mean to 10% of the
>mean. In all the examples I worked out a .1 or .2% increase in the rate of return
>completely dominated that.
Do you seriously find this exercise helpful? Couldn't you just as
easily back out the (von Neumann-Morgenstern, I presume) utility
function you need to get an introspectively plausible answer? In other
words, if you feel nervous with a SD of 20% of the mean, could looking
at utility functions really make you feel better about it?
--
Prof. Bryan Caplan
Department of Economics George Mason University
http://www.bcaplan.com [EMAIL PROTECTED]
"He lives in deadly terror of agreeing;
'Twould make him seem an ordinary being.
Indeed, he's so in love with contradiction,
He'll turn against his most profound conviction
And with a furious eloquence deplore it,
If only someone else is speaking for it."
Moliere, *The Misanthrope*
