Nt backtest features

[nicknamed]

Hi,
I know that in v6.5 NT backtest only for close prices, but when I trade for Moving average price of certain frequency, will that be possible in NT7?
N.

[NinjaTrader_JoshP]

nicknamed,

I am not sure what you mean by "certain frequency". Can you please clarify? Thank you.

[nicknamed]

specify

nicknamed,

I am not sure what you mean by "certain frequency". Can you please clarify? Thank you.

In NT6.5 backtest works on bar close prices. but I do not trade for bar close prices - if there is algorithm that calculates for what prices it is being traded (like moving average with period 2 - say SMA2) - would it be possible to get from NT7'backtest results corresponding to that SMA value? Well, for higher timeframe NT would need to calculate from minute data specific values of SMA and calculate which was first one (or the worst spossible) to return a value for my statistics.

This is not a simple thing. I hope I did explain it better, if not, write me again. please.
Thank you

[NinjaTrader_JoshP]

Guess I still don't follow you. You can always access the SMA value as well as OHLC. The only thing not available is intrabar information.

[nicknamed]

spec2

OK, one question in the beginning before any explanation: may I set in NT65 the backtester to return entry/exit prices for SMA(3,median) when that crossed a bar (since it does not happen every bar) or any other value different for bar close values? Just to find a value that might corerspond with real trade prices...

So back to my explanation:
I know that I can access those values but since they are not intrabar they are not reliable. imagine a system that trades for SMA2 when something happens. then I want to backtest this system and need the script to give me trades (prices of entry and exit) that were for that SMA2. On 1 minute it cannot be done closer than SMA(2,close) , but on higher timeframes say 30 mins, we can take bars minute after minute when waiting for the price to intersect SMA2 on 30Mins and keep calculating the upper SMA2 on 30 mins, then we might find a single minute during which We obtained the price at sma2 on 30mins and entered the trade. well, the minute, as we cannot get inside, might draw -depending on its volatility- different values of SMA2 on 30min timeframe (the timeframe of trading), so the backtest would take the worst value (say in the beginnning of a minute the SMA would be lower for long entry so the profit might be higher in the end), I mean if there was also much worse value of SMA on 30mins (the execution price) then the backtest would take the worse price of a minute that would cause an entry.

I have such a system and I am looking how to backtest it as precisely as possible, since current backtest does not return reliable values.
Thank you

[NinjaTrader_Bertrand]

nicknamed, in backtesting the entry is executed on the next bar open, to simulate getting filled intrabar, you can review this sample for a concept already usable in NinjaTrader 6.5 - http://www.ninjatrader-support2.com/...ead.php?t=6652

If this is still not fine enough for you, try recreating your higher timeframes calcs from a lower and finer chart, thus giving you more control.

Both can be done in NinjaTrader 6.5 through the use of a multiseries strategy - http://www.ninjatrader-support.com/H...struments.html

In NinjaTrader 7 this concept is extended to indicators as well...

http://www.ninjatrader.com/webnew/NT7/NinjaTrader7.html