> Here are some notes I took from the Scientific American:
>
>> 97: An unpublished study by Jens Carsten Jackwerth of the London
>> Business School and Mark E. Rubinstein of the University of
>> California at Berkeley has shown that traders' own rules of
>> thumb about inferring future stock index volatility did
>> better than many of the major modeling methods."
I'm interested in what you think the implications are of this statement (which I agree with, btw). The history of computation is full of such examples; but is this an indictment of the failure of modelling to accurately capture things like "instinct" and "experience" or just a portrayal of giddy managers hoping (against hope) that computers will replace the bullys you find on trading desks?
I think it's important to separate the use of quantatative methods in trying to understand risk (such as JP Morgan's Risk Metrics) and agressive speculation based on some arguably unscientific future-predicting model (cf neural networks) on a proprietary trading desk. You seem to have conflated the two in your choice of quotations from the article; modelling of interest rate or currency risk for use _as a hedging tool_ *is* much more accurate and successful today than methods used in the past. It is perhaps because of this success that the temptation to use the same methods for the other side of the equation is so high. But then you see things like "Oh, I only hedge when I think it's likely it will go down" -- which turns a hedging activity into speculation. "Say, what's the liklihood that we have a flood this year? Let's skip the insurance payment and go on vacation!" ... whee!
It's especially depressing to see non-finance-sector companies using derivatives in _any way other than hedging strategies_. It's like GM coming out with a line of action figures -- it's just not in their business plan to speculate on anything other than next year's car models. It's no wonder we're seeing these kinds of activities blow up.
But back to the quotation: yes, the human brain is still King in terms of predicting the future. Speculation involves predicting the future; hedging involves *removing the need to speculate* about the future, removing an exposure to future uncertainty.
Totally different animals.
/jordan