backtesting renko ranges ?

[NinjaTrader_Brett]

Marco,

You do not understand, you do NOT have to go to market replay. You need to enable intrabar granularity on your backtest. This is done with this link let me know if you do not understand it.

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

-Brett

[marcow]

Brett,

I did some tests in the past with adding a secondary dataseries for intrabar granularity, but came to the conclusion that it slowed down everything to the point it was unworkable anymore.

But I'm willing to give it another try.
So let's assume my main dataseries is Renko 4 and to get correct backtest results I have to add a secondary 1 Tick dataseries ?

[NinjaTrader_Brett]

I did some tests in the past with adding a secondary dataseries for intrabar granularity, but came to the conclusion that it slowed down everything to the point it was unworkable anymore.

Thats the problem, the more data points (Bar Data) the more time it takes to back test. Backtest is a delicate balance between accurateness and giving up some of that so you dont have to wait ages for a test to complete. Each trader needs to find the balance that works for them.

1 Tick series is going to add a lot of time to the backtest, I might suggest going with a 100 tick secondary series foe example. Would get you the more accurate results without so much of the performance impact.

I hope you can now see why it is this way in the first place.

-Brett

[firstbrain]

Hi Brett,
since I'm having the same problems with the renkos and don't quite understand what u try to tell us I have the following 2 suggestions:
1. why r u tellung us that it is better to take the open of the next bar instead the close of the actual bar. saying this 3 times and saying u won't change this doesn't make this argument correct or more viable. but if it is really better to use the open of the next bar, please tell us why!
suggestion: just use Close instead of Open only for renkos - this solves the whole issue (regarding speed and accuracy), since u won't need tick granularity or whatever u suggested. at least it would greatly enhance the results without compromising time constrains since the results we r getting at the time being are basically worthless.
2. i would suggest that u at least should do some kind of popup (as long as u can't resolve the problem) to warn inexperienced users of your platform if they trade "live" off of the results they get with your platform which may be totally incorrect and so could cost them money.
Hope I could help a bit to solve the issue. Pls correct me if I'm wrong.
Thx,
firstbrain

[NinjaTrader_Brett]

Thanks for the feedback I will forward in and happy trading.

-Brett

[precogtrading]

Hi, well this is 2017 (many years after the original post) and I just went through the hassle of finding on my own about this "feature" of renko bars. I am wondering if any of you have a newer solution or workaround for this issue. I am not very happy about adding a secondary time frame to my strategy as this would mean a lot of work and time for the backtests and optimizations.

Thanks in advance.
Alex

[NinjaTrader_Brett]

We introduced a new feature in NinjaTrader 8 for exactly this case, which is setting the 'Order fill resolution' to 'high'. This automatically adds in a the needed secondary series without any effort on your part if your using bars which 'hide' the true market action like renko's do.