Backtest problem: Entry and stop loss within one bar

[DH008]

Hi!
I am developing a strategy for a client which is trying to catch big moves.

The backtest results are complete wrong because whenever the entry price and the stop loss price is within the same (one minute) bar the backtest executes the stop loss.

I know that this behavior is intended but in order to get my clients backtest to work I would need a workaround fot this!

Any idea how to fix this?

Thanks in advance!
DH

[NinjaTrader_RyanM1]

Hello DH,

To work around this you have to move away from the Set statements and code a multiseries strategy. The secondary series should be smaller than the primary, and you use the advanced overload for methods ExitLongStop() and ExitShortStop() that allow submission to a specific series.

http://www.ninjatrader.com/support/forum/showthread.php?t=6652

[DH008]

Ryan,
Thanks for the quick reply!

The problem is, I am working on 1 minute bars which is the smallest time frame I have....

Any undocumented features?

[NinjaTrader_RyanM1]

There are no undocumented features here and generally we are unable to provide support for items outside the documentation.

Alternatives:
Adding a series smaller than minute - Tick.
Start with a wide stop loss that is submitted on the entry bar, and then narrow this after one bar.
Use market replay with fast playback speed (can go up to 500x) to simulate strategy behavior in real time conditions.

[DH008]

Well, that's not helping me.
Most of the backtests deliver wrong results with this feature.

Generally, that's a known problem in back testing.

Other systems have optional functions to prevent a stop loss event in case the stop loss is within the entry bar and the entry bar points in the direction of the trade.

Maybe you could implement this feature in the next release. And, it would be very helpfull to have a report pointing out this special situations.

I am doing back testing since quite a while, with different systems for different clients and this is an issue which pops up pretty often, and, it can change back test results dramatically.

best regards
DH

[NinjaTrader_RyanM1]

Thanks for your feedback, DH. We have this is a suggestion with ID # 688.