First of all I am not into total automation. I want to end up with clear charts that give a clear signal on a long term then a short term- arrows that through backtesting will give me the best chances of success.
TO do this I need to pretend I am automating and set the strategy to go long or short under particular circumstances.. This is only for the sake of the strategy analyzer to tweak the best out of it in order to then replace the entries arrows.
I am only a wizard user and so have limited space for parameter inputs.
This is my thinking and I would appreciate someone else chipping in their two bobs worth.
I notice that when the macD crosses above the average- price goes up and when it crosses down price goes down.
SO I set up a wizard strategy with the various user defined inputs as MAcD fast, slow, smooth. THen I backtest it. Then I optimize it through the genetic optimizer with as big a chunk of historical data that I can muster... currently 3 years.
I make note of the best tweaks and then move on to another indicator and do the same.
Eventually I will have a system of individually tweaked strategies that over a long period of time have proven successful.
Then I build one comprehensive strategy incorporating all of these parameters and tweak that according to input series.
Is that anything like how people are working?
thanks
raycam