Hi,

The code is found below.

I use PRT 10.2, so I haven’t checked the tick by tick back-test.

The spread is included in the simulations. What is the “0 bars issue”? Perhaps that could be it since it closed on the same candle.

// Definition of code parameters
DEFPARAM CumulateOrders = False // Cumulating positions deactivated
DEFPARAM FlatBefore = 071500
DEFPARAM FlatAfter = 215500

indicator1 = Average[34](close)
indicator2 = Average[36](close)
NoC = CurrentMonth - 1  //Adjust for February
//indicator3 = Stochastic[14,3](close)

// Conditions to enter long positions
c1 = (indicator1 CROSSES OVER indicator2)

IF c1 THEN
BUY NoC CONTRACT AT MARKET
ENDIF

// Conditions to exit long positions
c2 = (indicator1 CROSSES UNDER indicator2)

IF c2 THEN
SELL  AT MARKET
ENDIF

// Conditions to enter short positions
c3 = (indicator1 CROSSES UNDER indicator2)

IF c3 THEN
SELLSHORT NoC CONTRACT AT MARKET
ENDIF

// Conditions to exit short positions
c4 = (indicator1 CROSSES OVER indicator1)

IF c4 THEN
EXITSHORT  AT MARKET
ENDIF

// Stops and targets
SET STOP pTRAILING 2

I know it looks weird with the narrow span on the MA’s and tight stop loss, but since the results were great, I couldn’t resist trying it live.

Cheers,

Tobias

Hi,

Thanks for your replies. I thought the trailing stop at 2 points seemed funny, but that was the result I got when optimizing it. I’ll adjust it and try again.

Henrik, do you have any explanation for why the stop is 10p instead of 2p?

So basically, the backtest can’t be trusted in PRT 10.2 without tick by tick backtest?

Tobias