Volatitily as social flaring

J. Barkley Rosser, Jr. rosserjb at jmu.edu
Fri Feb 16 15:04:53 PST 2001


Doug,

Frankly I am in a bad mood today and just resigned from another list because I got pissed off at the behavior of the listowner. No, I am not about to do that here for the moment. I don't think I'm in as bad a mood as Mad Max, but maybe...

OTOH, I have encountered your general skepticism about applications of math before, and have on occasion seen you drive them into the ground, an act I am not interested in participating having you repeat. That does not mean that your objections are not well grounded ultimately (although I disagree with them). My condescending reference to you poring over tables of numbers was partly to point out that there are a lot of people, I suspect a few on this list, who view tables of numbers as being as meaningless as any other kinds of math, which tables of numbers are. (I could go on about all the difficulties involved in gathering data and what really lies behind all those numbers, but let us not get too boring.)

Actually, saying that leptokurtosis implies that markets are overshooting reasonable valuations on the way up and the way down is not necessarily true. There are financial economists out there, e.g. Peter Garber, who justify what we see in the markets as being perfectly reasonable on various grounds. I don't happen to agree with that, however. Leptokurtosis, although long and perhaps scary to those who have not had math or stats, is far more precise in its meaning than a lot of the words that appear in some of the works that you take very seriously. It does translate as "fat tails" especially in comparison with normal distributions, which might not be associated with "reasonable valuation targets," although plenty of models assume that.

Math models in general do allow one to clearly see or state or understand the assumptions that bring about certain results and how they are brought about. That is the case in the specific model that has been referenced. I do not assume reasonable valuation by agents but a form of imitative behavior. What that leads to can then be seen precisely.

If you are implying that the only use of such analysis is to make money, well, I am not going to do that, although there are plenty of people who are using such approaches to try to do so. But, frankly, I thought it would be preferable to make an ironic wisecrack rather than go through all this stuff again.

And, just for the record, I am not one of those who says that "math is all there is." A check of my website will show quite a few papers that have neither equations nor figures nor even data in them. In this matter, I swing all ways (heck, I even do econometrics on your precious tables of numbers sometimes, too). Barkley Rosser rosserjb at jmu.edu http://cob.jmu.edu/rosserjb -----Original Message----- From: Doug Henwood <dhenwood at panix.com> To: lbo-talk at lists.panix.com <lbo-talk at lists.panix.com> Date: Friday, February 16, 2001 5:36 PM Subject: Re: Volatitily as social flaring


>J. Barkley Rosser, Jr. wrote:
>
>> Obviously you are right and I promise to clean
>>up my act and reform my wicked ways. Clearly
>>math is an illusion. Examining the details of how
>>a set of equations depicting investor behavior
>>generate certain outcomes over time via computer
>>simulation is as empty and meaningless as spending
>>a lot of time poring over tables of numbers.
>> Henceforth, let us both agree to only write our
>>ideas about economics in poetry, preferably Romantic.
>
>Might you descend from this summit of condescension and explain just
>what this accomplishes? It's pretty well established that markets
>overreact - which, as you explained this morning, can be described as
>"leptokurtosis," which sounds a lot more rigorous than saying that
>markets overshoot reasonable valuation targets on the way up and on
>the way down. What would a computer simulation tell you? I can
>understand if you're trying to make money or sell market projections,
>but what, scientifically speaking, would be gained by this exercise?
>If I were given to reckless psychoanlysis rather than asking
>questions, I might suggest that your "empty and meaningless" comment
>actually expresses your own anxiety that the exercise is empty and
>meaningless - safely buried in a defensive irony - but I'm not given
>to such, and I'm really asking questions.
>
>Doug
>



More information about the lbo-talk mailing list