So you have your EA, logic works and you need to get better values for time frames, thresholds, SL. TP whatever you may think will give you more money..
Next thing you would do is a back-test with a year or more data. Now here is the trickiest part, depending on the time of the test results are different. You can get 2 different results on 15 minutes interval.
It looks like Empty4 will use variables from that moment in back-testing. So a back-test on Asian Hours time will have no relevance for an EA designed to work on London Hours.
Empty4 Backtesting and Optimisation
-
Mariusz
- Trader
- Posts: 19
- Joined: Sun Dec 11, 2011 4:29 pm
Re: Empty4 Backtesting and Optimisation
jb86,
That is due to the fact that crap Empty4 does not store the historical spreads and uses the current spread for backtesting.
It gets bizarre if you do backestiting during the weekend as crap Empty4 takes the last spread from Friday (usually very high) and the results are usually much worse that in real...
The way to go around is to use soft that 'freezes' spreads for backesting like: http://www.mt4i.com/spreadcontroller.aspx.
That is due to the fact that crap Empty4 does not store the historical spreads and uses the current spread for backtesting.
It gets bizarre if you do backestiting during the weekend as crap Empty4 takes the last spread from Friday (usually very high) and the results are usually much worse that in real...
The way to go around is to use soft that 'freezes' spreads for backesting like: http://www.mt4i.com/spreadcontroller.aspx.