1 Topic by damir (edited by damir 2010-03-10 12:43:08)

  • damir
  • New member
  • Offline
  • Registered: 2010-03-10
  • Posts: 1

Topic: Some False results

First of all I wanna thank you for effort.

I tested some basic things (1 day TimeFrame EURUSD open price), that is with price move and with constant take profit and stop loss nothing more and on 100 results given around 10% are false where is profit it should be loss only one was opposite. All this was under pessimistic scenario and while checking i didn't find any reason that data could be interpet different just maybe in one case.

Note: I didn't test to much just 100 results

  • zenoni
  • zenoni
  • Licensed Member
  • Offline
  • Registered: 2009-10-25
  • Posts: 339

Re: Some False results

This seems to be an important issue. Please give more details about the strategy and conditions you used.

damir wrote:

First of all I wanna thank you for effort.

I tested some basic things (1 day TimeFrame EURUSD open price), that is with price move and with constant take profit and stop loss nothing more and on 100 results given around 10% are false where is profit it should be loss only one was opposite. All this was under pessimistic scenario and while checking i didn't find any reason that data could be interpet different just maybe in one case.

Note: I didn't test to much just 100 results

3 Reply by damir321 (edited by damir321 2010-03-17 01:23:41)

  • Registered: 2010-03-01
  • Posts: 12

Re: Some False results

I'll write something later when i see more clearly it seems that i miss some data so i have to look more deeply

  • Registered: 2010-03-01
  • Posts: 12

Re: Some False results

Well i'm new to the program so i didn't know about bar limitation so the problems are solved with editing config and cutting bad data but i have another question, those bars that on the lowest time frame available where on the same bar is open position and logically it could be take profit in the same bar and they are mark ambiguous and if they aren't take profit in that same bar later they would be on the stop loss on the bars ahead, why are they mostly (what i saw 98% of the time)interpreted like profitable in pessimistic scenario and they don't change in the optimistic scenario (how could they change cause they are already optimistic in pessimistic scenario)

Another question I downloaded some data from 2001 to 2009 and edited config lets say to 500000 bars limitation or higher but when I open FSB my intrabar scanner only shows 1 minute and 1 day  calculated (1day strategy) why is that.

  • Registered: 2009-12-16
  • Posts: 19

Re: Some False results

There's definitely an issue with the results of the intrabar scanner. 

For instance if you take an opening position in 1 hr timeframe, go long at Heiken Ashi low and then assign Close and Reverse exit, you'll see amazing results under both normal optimization and also the intrabar scanner with 1 minute data.  If you look at the backtest, entry points are all at the candle's high and low...or HA high low. 

However once traded live, entry points open in the middle of candles.