The OP's discussion of idiosyncratic high-frequency noise is happening in the context of an automated system using a computer to make the decisions, no "psychological part" needed. I assume that because this is the "Automated Trading" forum. Since backtesting is the absolute core of an automated trading system your comments are quite extraordinary. In other posts it sounded to me like you have methods that could be mechanically traded. That would also mean they could be backtested. I take it you would not agree?
Dilemma in modeling idiosyncratic high-frequency noise
- Thread starter bluelou
- Start date
Hmmm... When the answers just create more questions best to just stop. Sorry to the OP for getting a bit off topic. I must dash off now to explore what I have to do to earn the official title of "master programmer".
ProfLogic,
What part of your approach to trading is automated? How did you find out what your edge was or figure out what to automate if you didn't look at historical data?
Regardless of whether you casually observed the data in real-time or you looked at the historical data systematically (backtesting) you've still exposed yourself to a data snooping bias.
-bluelou
What part of your approach to trading is automated? How did you find out what your edge was or figure out what to automate if you didn't look at historical data?
Regardless of whether you casually observed the data in real-time or you looked at the historical data systematically (backtesting) you've still exposed yourself to a data snooping bias.
-bluelou
A computer can be taught to read the oscillations and tell when a trade is setting up, oscillation to oscillation and then, bar-by-bar, finally executing the trade.
The problem is the interface with the trading platform. This is why I currently have the process automated to tell me when a trade is setting up, based on oscillations, out of the 1000+ contract months I monitor, so I can fine tune it down to a handful of charts. At some point in the near future, my programmer will complete a macro that will run the program automatically each morning to tell me which charts are potentially setting up trades that day. I can then take the trades manually. The step after that is integration into Ninja for the execution.
Tradestation would be great because the the platform is built in but Tradestation's Constant Volume/Share Bars aren't accurate. They aren't capped and they refuse to take the few hours to fix the problem.
To answer your second question, real screen time. Five years of watching the charts, eight plus hours a day for 5 years. Nothing casual about my research, I assure you. If you take a lax approach to a problem . . . the result will mirror the effort. This gave me the information that price moved perfectly and logically so it could be programmed. I leave the programming to those that can.
By the way . . . you can't program variables where the variables are infinite and when you use minute, tick, equi-volume, momentum or range charts, all outcomes are infinitely varying.
I don't know about the days when Wang Labs ruled the roost but these days programmers actually prefer to be called "Software Engineers". I have always thought of programming of more of a craft than a science so I believe one does eventually become a "master" of the craft to some degree. Problem is in most circles I know of one would be laughed out of the building if they printed "Master Programmer" on their cards (but secretly like it 
).
).I personally agree. I went down that road unfortunately and found cool profitable systems but then discoverd they were impossible to trade with real money.
Why? You can't "feel" the pain of a draw down on a historical chart until you see it on your trading account.
Where I only slightly disagree with your post is this:
After a person has real life experience trading real money, doesn't their historical testing skills become exponentially better (provided they have the right tools--like any skill)?
They do. It's because the equity curve and MAE and MFE's etc. all take on a more meaning after you suffer through real great MFE's that become losses or draw down periods.
Real life experience forces you to think about money management in your and build that into your strategy.
It also forces you to raise your "tool" standards towards true market simulation rather than just historical testing.
Just my 2 cents.
Wayne