The problem is predicting behavior in the money markets using conditional probability. One theory, though not a necessary one, is that the path of prices directly reflects the underlying psychological states or utilities of the participants. There are non-random patterns that can be identified that are predictive of future price paths, just as three shouts might lead to a hit in a coercive family. It's the scientific approach. Qualitative models are used to create statistical models. We have looked at various models of frustration aggression in the markets, war, strategy, sports, survival, revulsion and release, evolutionary adaptive models, game theory, physics, mechanics, electrical theory, hunting, Tversky and Bayes. I believe that you have some theories about war and broader human tendencies using a statistical analysis. The markets and market data capture these broad human characteristics in an amazing way and the data is there ready made.

Jerry Patterson comments:

It had never occurred to me that the frustration aggression theory might have some relevance to behavior in the money markets. I was intrigued by your approach of applying conditional probability analyses to the problem. My understanding of economics is that one of their major problems in prediction is that none of the parameters in the prediction models contain terms based on measures of human behavior.

This comes close to laying out a problem that would interest a psychologist. In fact the Nobel prize in Economics last year went to Kahneman and Tversksy–two psychologists who spent their life calculating conditional p values describing risky choice.

Russ Sears adds:

From last week's performance, I believe the question for 2007 will be "how does the market perform 'when good news is bad'?" Psychology mixed with counting should get you far.

Vincent Andres mentions:

One prior side of the "patterning" job is quantitative statistics–finding non-random behaviors. Another side concerns the "whys" of the behaviors of the "mass market" entity. Concerning this second point, I found the following books of K. Lorenz very penetrating and enlightening: The Foundations of Ethology (For example, you should read II.I.7, which discusses stimuli levels reduction. The French version of this book can be found here.), Studies in Animal and Human Behavior (Many thoughts also concern fear, aggression, frustration, etc.)

I believe that the non-randomness of the mass markets, for example, its behaviors as an entity, emerges precisely at common denominator points of all its individual human components. Common denominator points are not cultural/sophisticated ones, but primitive, for example, animal ones. That's why ethology, and especially the work of Lorenz, may concern those of us interested in the "whys."

Trying to answer the "whys" may be useless, but it could help to define/find patterns we won't think about otherwise (and understanding, at least trying, cannot harm).


Resources & Links