Mar

7

A few years back I read Wolfram's A New Kind of Science. I have been thinking about snow and looking forward to some heli-skiing in Alaska next week while spending long hours watching the market day and night. Wolfram's thesis is that simple binary rules for cellular automata in computer-generated binary or trinary functions develop into patterns as they branch out in an iterative fashion, often random in appearance, but with astonishing regularity. Quite amazing regular patterns, symmetrical patterns, emerge.

One of his theories is that natural cellular development is often binary and such a basic mechanism leads to gross formations with bilateral symmetry such as a hand with five fingers, a leaf, shells on the beach, starfish and a host of others. Applying the ideas to crystal formation, such as snow crystals, a simple branching mechanism tends to create symmetry in its patterns, though varied to infinity within its randomness, but more than random nonetheless, due to the basic binary rule at the heart of the creation of the crystals.

The market bid-ask is a simple binary function which, when iterated, develops the many price patterns in the historical record. The interesting application is the appearance of more than random occurrences of regular and symmetrical patterns. TA practitioners have proclaimed this for years. But might there be a quantitative manner of deconstructing and predicting the formation of a pattern before it completes, in a rigorous predictive manner?

Though he could develop the patterns from a basic beginning formula, Wolfram's greatest question and unsolved problem was that he was unable to deconstruct the basic formula from the developed pattern. But his main query for further study was, if these patterns could be developed from simple basic iterations of a simple rule, then why can't they be backwardly deconstructed?

In the market we know the rules; we see the patterns. Why can't they be deconstructed, categorized, and thus be predicted in a more rigorous manner than the TA practitioners use? There is no question in my mind that some TA rules have grounding and can be used rigorously for prediction. The Popperian problem is to state it in a falsifiable form, in a manner that can be quantified for testing.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search