Why my backtest result using the same EA for 2 brokers is totally different?
-
Hello
Please, I need some help to understand what is going wrong with my EA built with fxDreema when I backtest or optimize in different accounts (MT5). Because the results are different from one account to another.
I have optimized the EA in a MetaQuotes account. I have chosen the paremeter set that is giving a good consistency through 10 years (in sample) of backtesting and 1 year (out ofsample) of forward testing. The result is this
It seems a good EA. But, when I shift to my broker's demo account using the same paremeter set, the results are totally different
It seems that the EA's behavior in out of sample period is close for both accounts. But for me, if the EA does not have the same behavior as much as possible for both time periods (backward & forward), it is not a consistent system and I will discard it.
Could be anything related to the historical data quality?
Thanks for the help
-
history which is not exactly the same for all, using candle ID 0 for signals, using repainting indicators, using different (mostly small) timeframes ... many reasons ...
-
@josecortesllobat I lived same problem before. So after I find where the my issuee. İts caused 4 and 5 digit broker. Maybe you got it same problem.
-
Hello
Thanks for your answers.
@casper Both Metaquotes and my broker have 5 digits. I have also built the EA using 5 digits by setting 0.00001 on "Project Options" for most of the pairs that are using 5 digits.
@miro1360 As I know, MT5 downloads historic data from its servers not from the broker's server. But probably I am wrong. The EA is working with the same result in my broker's demo account and also in MetaQuotes demo account.
I appreciate the help.
Regards
Jose
-
this is because of account types, broker types, leverage, spread , slippage, time delay . etc.
every broker has their own algorithm . also the broker time is important as well -
Noticias
