---------------------- Forwarded by Vince J Kaminski/HOU/ECT on 07/11/2000 
02:59 PM ---------------------------


Tom Arnold <tarnold@finance.lsu.edu> on 07/11/2000 12:32:27 PM
To: Vince.J.Kaminski@enron.com
cc:  
Subject: Re: EFA meetings


Hey Vince,

Thanks for your reply.  I'll see what becomes of the session and keep you
informed.  As to the paper, Tim Crack and I have a revised version of the
paper I gave you.  We have since found out that by using certainty
equivalence, our model is more robust.  For example, if one has an asset
pricing model that incorporates mean, variance, and skewness (Harvey and
Siddique, JF June, 2000) and a binomial model that incorporates mean,
variance, and skewness (Johnson, Paulukiewicz, and Mehta, RQFA, 1997), our
model allows you to price options under the real world measure.  The
benefit is that one can take all of the model parameters from historical
data that is non-risk neutralized.
   From a pricing perspective, there isn't a tremendous benefit in a
mean-variance world (variance stays the same in risk neutral or risky
measure).  However,in the mean-variance-skewness world, there is a benefit
because we do not believe (although we're still hunting down an appropriate
cite) skewness is the same under risk-neutral and risky measure.  Given we
can only measure the skewness in our risky world, our model becomes much
more significant.
   I would certainly appreciate comments on the version of the paper you
have and would also pass on the new version of the paper if you would like
to see it.

Thanks again,

Tom