Thread: Golden Eagle Ea
View Single Post
  #7 (permalink)  
Old 12-20-2005, 11:20 PM
kasmodiah kasmodiah is offline
Junior Member
 
Join Date: Nov 2005
Posts: 15
kasmodiah is on a distinguished road
Quote:
Originally Posted by profisharm
HERE IT IS.CHECK THE RESULTS OF BACKTESTING.THESE ARE VERY INTERESTING,BUT I THINK,IT HAD A MISTAKE.MAY BE,ANYBODY WILL TEST IT?BEST TIME FRAMES ARE H1 ON EURUSD and GBPUSD.
Ha Ha, you have 59% model quality (right top corner of your report) and you get all itchy about not posting the mq4?

I am actually to tired for this but I want to be fair and explain this:

Why is Dataquality important for the evaluation of backtesting results?

If you have less then 90% model quality, you have "odd gaps" in the data. Sound like being just too fuzzy, eh? No, its more. Imagine you put up a trade with SL, TP and a nice entry Price. At the moment where in real life the SL and/or the TP would hit it, or maybe just the exit signal you have a gap in the data. So the backtester does, what it can do. It pretends that there is no gap. So a long trade which started at 1.25, and hit every SL way down to 1.10 and went back again to 1.35, giving you all kind of exit long signals and enter short signals, gets thus simplified to 1.25 -> 1.35 didnt even see a SL => voila you made a great gain, which never actually happened.

(Damn thats a lengthy explanation I admit).

What I want to say, when you have a gap in your data, you can consider greatest crap being the alchemy of forex trading, or - even worse - oversee a great system, because your dumb data, does not correspond with it.

So give youself a chance, check you model quality. I am Mr. Backtest and I have spend hours just get the data for the 4 majors with 90% model quality back to june 2004, so why should you get itchy about a backtest with 60% model quality????
Reply With Quote