Friday, March 30, 2012

Using Perturbation Theory to Build Trading Models and Systems

I have recently finished reading Mr. Brian Greene’s book, The Elegant Universe. It is a fascinating book about string theory. One of the concepts Mr. Greene discusses is how physicists use perturbation theory to advance their knowledge and theoretical ideas.

Mr. Greene says that the mathematical framework of string theory is so complicated that physicists have to use approximate solutions and calculations, while initially ignoring some details. Later, these solutions and calculations are further refined as more details and knowledge are systematically included. The perturbative process can also be used to build trading models.

I had no idea of perturbation theory before reading this book, but I now feel it describes the process of how I created trading models. The stock market is also complicated and difficult to understand (probably more so then string theory). Initially, I began to systematically organize some variables, ideas, and concepts, while leaving others out. I began to create some general ideas, theories, and approximate solutions to trading. Later, with further testing, experimentation, and pain (the great teacher), I was able to increase my knowledge of how the market works, and I began to systematically include more details and concepts. This led to further refined ideas and theories which eventually became my trading models.

But here is some advice and a word of warning to traders out there—the key is to use a perturbative approach up to a point. The irony is that the more variables added to “refine” and “improve” a model, the more likely it is that the model will breakdown.

Thursday, March 29, 2012

Mr. Larry Williams and Mr. Niederhoffer's, Daily Speculations

I like to read different blogs in order to educate myself about a number of issues and topics. One blog that I like is Mr. Niederhoffer’s, “Daily Speculations.” This blog contains some interesting topics submitted by a variety of individuals. The other day I read a comment by Mr. Larry Williams. He said,

“The other day I heard somebody say:
"Assuming the future behaves the same as the past, I reason that this way makes my funds efficiently used".
I wanted to say, my experience is that the past is never like the future so we waste valuable time and skills on a false postulate.
As I see it, it is better to have a core strategy to deal with equity drawdowns, etc –based on logic–as opposed to a strategy based on the past real results or back tested as that is for the most part a make believe world since it never happens quite that way again.”

Most system traders backtest their ideas and models in order to have some confidence to trade into the future. As a system trader I always think about the future and if it will resemble the past or not. I wonder if the ideas behind my models will ever fail going forward. Mr. Niederhoffer likes to speak about ever-changing cycles in markets. These ever-changing cycles may eventually cause the breakdown of trading models. Perhaps, this is true.

Mr. Williams has spent much of the past teaching traders to backtest ideas, etc. He has made money teaching and writing books that include backtesting, etc. Has something changed? In addition, I believe Daily Spec removed Mr. Williams initial comments (or perhaps I cannot find them), but I can tell you that Mr. Williams had a much different tone in that post regarding the use of backtested info and using it to trade with.

I agree with Mr. Williams about having a logical strategy to deal with drawdowns, but backtesting is not a waste of time.

Friday, March 23, 2012

Our redesigned website-

We have spent the last few months redesigning our website, TRADINGXYZ.COM. Please review it when you get a chance.