I’ve completed about half of the book on Einstein, and it is very apparent that he prefers deductive reasoning over inductive reasoning. Simons, in 2000, made the opposite argument regarding his Medallion Fund:

We don’t start with models. We start with data. We don’t have any preconceived notions. We look for things that can be replicated thousands of times. A trouble with convergence trading is that you don’t have a time scale. You say that eventually things will come together. Well, when is eventually?

This identifies a divide between the methods of trading strategy creation.

First, you have those who believe that trading methods should be created through the analysis of existing characteristics. Behavioral finance and much of market microstructure theory are based upon this type of reasoning.

Second, you have those who believe that any pattern or anomaly derived from data is just as substantiated. Medallion’s historic returns are based upon this type of reasoning.

I thought this was interesting after reviewing ASTA, which seems to have capabilities of supporting both approaches. Perhaps, the solution to the question of which method is more appropriate is based upon what the trader believes they are looking for. Einstein was looking for underlying laws, and from a prior post we already know that Simons doesn’t believe in underlying laws in the financial marketplace, even if the principles that underpin his systems don’t change.

[Update] This post brings to mind a newsletter I received a number of months back from Mike at Breakout Futures. You can access the main article here. He provides simple code for TradeStation that will develop automated systems for the user, based upon the type of price pattern system the trader dictates.

While I haven’t used this type of system, I think it warrants further review with caution - it is likely similar in concept, although more rudimentary, to those systems developed at aiQUANT and Neural Market Trends.

Personally, I have used deductive reasoning to develop frameworks for money management methods, and then tested data in a similar, inductive method, to create the details.