|
PureBytes Links
Trading Reference Links
|
I love this talk about mechanical systems :-) actually it was Steve who dropped a formmula on the list, oh, that was shortly after I joined this list about a year ago, that revived my believe in Mechanical Trading Systems. I thank all of you for your contributions!
In terms of developing trading systems (measured in years!) I am a "newby" however I have noticed that the application of trading systems appears to be subject to timing. When testing basic systems, with basic I mean 'not highly optimized', I often find that systems work extremely well for several years and then stop working abruptly.
I have been unable to discover the reason for the sudden success and/or failure of systems. Sometimes systems based on completely different criteria and formulae succeed and fail within weeks of each other.
My gut feeling is that this happens too often to be the result of being lucky or unlucky. So, my conclusion is that the market, trading methods, order regulations & handling, and who-knows-what is subject to rather abrupt "periodic" changes that may occur at 6-24 months intervals. For example the new regulations at the Amex and NYSE may have changed trading patterns...
So, this makes me think that the application of Mechanical Systems is subject to "System Timing", just like you time a trade.
I am exploring (in VBA, MS can't hack it) running several systems simultaneously and compare the equities for each, just like you would do with indicators, to give me System Entry and Systemn Exit signals. Since the systems I test give me signals at about 3-12 days intervals the system could be fairly responsive.
Failing to identify the key turning points in market behavior the key to success appears to be lucky System Timing - that is, unless you use equity timing.
I have tested quite a few basic systems (many spawned by formulae on this list) that would have made me rich had I been around at the time they started working. While 'I didn't make it' :-( I am SURE their were bunches of other traders who discovered the same formulae, at the right time, and who made fortunes. This thought keeps me plugging away at it...
I invite your comments on the idea of Equity based "System Timing".
Happy trading, Herman.
|