difference in perfomance between 1 day and 1 min

[margandi]

hi,

I backtested strategy and noted that the same logic and conditions wriiten in 1 day ( conditions for 1 day) and 1 min (conditions for 1day) are absolutly different. To whom to believe?

conditions 1 day :
if (Low[0] < Low[1]
&& Open[0]>Close[0]
&&( ATR_percentage(100,14)[0] >= Bollinger( ATR_percentage(100,14),2,30).Upper[0] )
&& NDX(260)[0]<60 )
{
EnterShortStopMarket(100,Low[0] - 0.05,name);
SetStopLoss(CalculationMode.Percent,0.0025);
SetProfitTarget(CalculationMode.Percent,0.01); }

conditions for 1 min:


AddDataSeries( Data.BarsPeriodType.Day,1);



if (Lows[1][1] < Lows[1][2]
&& Opens[1][1]>Closes[1][2]
&& NDX(BarsArray[1],260)[1]<60
&&( ATR_percentage(BarsArray[1],100,14)[1] >= Bollinger( ATR_percentage(BarsArray[1],100,14),2,30).Upper[1] )
&& alreadyTradedToday == false
)
{
alreadyTradedToday = true;

EnterShortStopMarket(sharesToBuyatr,PriorDayOHLC() .Low[0] - 0.05,name);
SetStopLoss(CalculationMode.Percent,0.0025);
SetProfitTarget(CalculationMode.Percent,0.01); }

}

in 1 min there´s just 1 trade per day permitted.
Also i changed indexing -1 day in 1 min conditins .

so run these 2 strategies in 1 day and 1 min and there are different number of entries , in different days… where is the problema? How can i fox it? and which perfomance shpuld i believe?

[NinjaTrader_Jim]

Hello margandi,

Thanks for your post.

You will want to make sure that the logic is being processed on the same data series. If the additional data series is added, we will want to use a BarsInProgress check to ensure the logic is being processed on that data series if we want to have similar results. Having the logic processed on a minute series and a daily series will have different results.

I have attached an example that can demonstrate how close you can get the backtest when having logic and orders submitted on a secondary data series. This strategy can be compared against the Sample MA Crossover strategy on a daily series. Please note that the final Exit On Session Close behaviors will be different since the primary data series is used to fill the order. We can note in the screenshots attached that the final Exit On Session Close trade has the same entry price for the trade but different exit prices.

We look forward to being of further assistance.