Imperfect World, Imperfect Stock Market

In "Misfire" (Cover Story, Sept. 21), the recent dynamics of the market have reinforced a fundamental principle of investing: high risk, high returns. Quantitative investing attempted to dislodge that longstanding principle with "high returns with little risk"--but failed.

Dick Chaiet

Coconut Creek, Fla.

I am amazed at the things computer simulations can do: land a 757, make an unstable aircraft feasible (with 40 decisions per second), operate a chemical plant. But a decent engineer might accept the notion that some things are not predictable or quantifiable, and those that are might require a tuned response. How exactly does one factor in, or respond to, mass hysteria, tulip mania, or any number of chaotic events? Perhaps there's a problem related to collateral skills. I would feel more comfortable in an airplane designed by pilot-engineers.

S. J. Kavcak


Your surprise at the failure of quant "rocket science" is a little disingenuous. The plain fact is that the sophisticated mathematical models for stock trading and hedging you describe are, by nature, corrigible--since however perfect markets may be, price systems always contain elements of randomness and perturbation. What mathematics can help traders to do is to establish optional strategies in a world of bounded rationality where conduct in markets reflects the way sane people ought to behave in a sane world. Introduce the insanity of market distortions due to errors of judgment (from federal or central bank authorities or firms themselves), and even models operating in the most perfect and information-rich markets fail.

Blaming the math is liKe blaming horoscopes for a bad day--whether the horoscopes had inspired you to leave the house or shelter at home, you might still have had a terrible day. Here at University of Oxford, we are putting together a diploma to help "quants" get it right more often; we don't expect to be the solution to market makers' headaches, but we believe better math makes for better trading in the long run.

Mark Gray

Kellogg College

University of Oxford

Oxford, England

Before it's here, it's on the Bloomberg Terminal.