"can make a profitable system unprofitable. In many if not most instances, these particular changes are quite susceptible to simple observation."
What I was thinking of at the time I wrote that are some commercial trading systems have been optimized on the last few years of data. They can achieve a back history of high profits, but when you use them on new data, they will take a significant performance hit. Hence they are just a cheap curve fitted systems. I have used all 20 years of futures data to build and test my system, so I have more confidence in it.
"discretionary traders typically take advantage of multiple filters of different types: Ideally, they'd like to be trading the most tradable tradable in the most tradable way at any given moment."
I'd like every advantage "discretionary traders" have. "discretionary traders" is a new term for me. Every trader has to have a set of rules. To me a system is just a way to trade a pattern. My point is that a book showing a few charts with Bollinger Bands, moving averages, DeMark Indicators, etc., does not provide a tradable system in itself. And even if it did provide clear buy and sell signals, I still want to see the system tested to make sure it produces a profit.
I've scanned through a lot of TA books at bookstores, and alot of what I see is just a few charts with indicators on them showing an uptrend or a downtrend. But what's really important is: can you make a profit with controlled risk before the trend ends?
I'm not looking to give anyone a hard time here, I'm just trying to get some of my questions answered. I've been trained as an engineer, and when I design or manufacture equipment, I do rigorous analysis using finite-element analysis, bifurcation and arc-length continuation studies, sensitivity analysis, etc., to make sure what I propose is going to work. If you were to build a bridge, would you start by finding a book with a nice picture on it, or would you get some numerical analysis packages and conduct stress and failure analysis studies?
I want to know the statistics too, not just technical indicators. |
