Renko Range, pull closes to python work around?

[OllieFeraher]

Hi all,

I joined ninjatrader because I really wanted renko bars.

It seems they are unreliable for backtesting because they use the open price and not the close price from my reading of this and associated forums.

Is there a work around?

Are 3rd party bars a viable solution and if so are any recommended?

I am more comfortable in python, would there be anything unrealistic or any issue if I just copied all the bars into a python dataframe then run my own tests on the closes there?

Do range bars have similar issues making them unbacktestable in ninjatrader?

L,P,H

Oliver

[StealthM93]

You can give ninZa.com Renko bars a try. It's a free download.

https://ninza.co/free-ninjatrader-indicators

Hope that helps!

[NinjaTrader_BrandonH]

Hello OllieFeraher,

Thanks for your post.

Standard Renko bars can be challenging for backtesting as the barstypes use functionality to remove the last bar in the data series and replace the bar with a new Open price. This is done with reversals and cannot be simulated historically because the bars are already built.

Testing may be done with the Playback Connection to fully simulate standard Renko bars with accurate reversals and order fills.

Playback: https://ninjatrader.com/support/help...htsub=playback

Some custom Renko bars aim to work around this limitation and often substitute a "fake" open price in the bar. This will result in Standard Order Fill Resolution estimating fill prices using OHLC values that are not 100% reflective of actual price action which can create discrepancies with results. High Order Fill Resolution should be used in cases like this, or the strategy should be written to submit orders to a single tick data series.

Please review the help guide document on the differences on real-time vs backtest (historical).
http://ninjatrader.com/support/helpG...ime_vs_bac.htm

Let me know if I may assist further.

[OllieFeraher]

If the strategy submits orders to a single tick series on the close of the candle will that be sufficient for accurate backtesting?

Are the renko bars I see in the general chart legitimate, if so can I use their closes in a backtest legitimately if I have a dataframe of their closes?

Kind regards

[NinjaTrader_BrandonH]

Hello OllieFeraher,

Thanks for your note.

"If the strategy submits orders to a single tick series on the close of the candle will that be sufficient for accurate backtesting?"

This may be sufficient for custom Renko bars from third-party providers.

For Renko bars that come default with NinjaTrader, the Playback connection should be used to test custom NinjaScript strategies to fully simulate standard Renko bars with accurate reversals and order fills. This can be seen in the help guide documentation linked below.

https://ninjatrader.com/support/helpGuides/nt8/discrepancies_real-time_vs_bac.htm

Please let me know if I may assist further.