Dean LeBaron
October 17, 1996

Something is wrong, very wrong. With all the effort to collect information, understand, and interpret it, we should be doing better. I'm not talking about whether active managers are a mite better or worse than the benchmark. Rather, whether we satisfy our clients' investment expectations and our own estimates of our personal capabilities. And we don't.

Why not? Two reasons, at least.

The first looks back. All of us now use quantitative tools. These tools, introduced a generation ago and in vogue today, are designed for crude math and limited machines. We make simplifying assumptions in our models simulating the real world. We assume that the real world is linear (as though a few units of one variable have an equal effect on another variable . . . over the full range of measurement/estimation). Not so. The real world is distinctly non- linear, like almost everything else in the social sciences. We believe in synthesis, and we study smaller and smaller bits-bytes of data expecting they will retain their characteristics when put back together into a total system. Never. And we usually ignore the feedback features of markets in which each moment is a unique and unlikely-to-be-repeated event. Market time is in bursts, quanta not continuous. It may move from regular to chaotic to random . . . time is distinctly not the stable tick-tock that an equally-spaced x- axis implies.

And second (but there are more), market study, more than most, suffers from the Heisenberg effect-the results we get are subject to the identity of the researcher more than the phenomenon being studied.

This combination is the reason why investment strategies based upon hindsight often fail. Researchers all discover what they wished they had known, and they put it in place. . . but the alpha that was there in backtesting disappears. (As an aside-and footnotes don't seem to go with this informal style-I have observed that the most successful investment managers select investment styles consistent with their personalities . . . they understand the style at its very core, not from the numbers.)

Aside from the obvious flaws in conventional management, supported by evidence, there are some new approaches coming along. The pattern is remarkably similar to developments at the turn of the century in physics when better laboratory equipment found that Newtonian physics failed to explain experimental results. Something was amiss, and two schools emerged . . . one trying to restate the old work, better and better, and the other to take radical new departures. And from the latter we have quantum physics. So, in my opinion, we are at the same juncture in investment management eighty years later.

One of the forward-looking technologies offering promise of better science and better results is drawn directly from physics. It is called complexity. Sometimes chaos or adaptive systems, evolutionary dynamics and even artificial intelligence, neural networks and genetic algorithms . . . all seem to crop up. The problem with these ideas is that, although they do not have the flaws of conventional practices, they are fuzzy, usually unsupported by repeatable performance attributes, and still undergoing modification. Against the pseudo-precision of old ideas, they seem experimental and flaky. But that is what being on the frontier is all about.

Let me list some possibilities and observations, most drawn from successful applications in the physical sciences and engineering. I think they make the inter-disciplinary crossover.

1. Look at time in new ways. Redo time series in a variety of ways-drop some periods that may be misleading, emphasize others like the more recent periods, look at volatility for clues of the relevance of periods.
2. Introduce instability to get more stability in a chaotic system.
3. Be aware of the influence of feedback loops. And measure the shortest term you can find.
4. Same inputs can lead to different outcomes.
5. Do forward-testing, use scenarios . . . the equivalent of thought experiments for the early quantum physicists.
6. Be conscious of the events at the tails of the distribution. They are usually unrelated to the measurements around the middles and are often more frequent than anticipated.
7. Question, always, causality.
8. Results always depend upon the point of view of the observer. Data is not knowledge but a point of view of the observer.
9. Small changes of inputs can produce different and large changes in outputs.
10. Look for the presence of strange attractors.
11. Keep the rules as simple as possible to understand their influence in a complex system.
12. Consider the presence of fractals in investment data.
13. A system behaves differently than the sum of the parts because of the characteristics of complex systems.
14. Propensity of systems to self-organize.
15. If investors are looking at the same history, a naive forecast from that history may be the least likely outcome.
16. Markets are inductive, deductive, and random, all at the same time. Many worlds.

And there are more. In subsequent articles, I'll take each and try to apply them to today's conditions.