Bars.GetDayBar(1).Time returning null

[drolles]

Dear NT Support,

I’ve solved the problem, but just in interest of all NT users and yourselves I thought I best post here.

I’ve just been working with GetDayBar. It had been returning null exceptions.

It was returning null exceptions. I’m running a multi-timeframe strategy with a focus on 60min charts with a secondary focus (i.e. Add(PeriodType.Minute, 1) of 1 min. GetDayBar was returning null when I was using it to track dates (i.e. currentDate = Bars.GetDayBar(1).Time). This would continue to return null object exceptions. However, when I added a third timeframe (i.e. Add(PeriodType.Day, 1)) the strategy ran fine and returned expected results.

Are we required to have a “Day” timeframe for GetDayBar to operate correctly? Is that expected behaviour?

Thanks and regards,

drolles

[NinjaTrader_Bertrand]

drolles, checking for null is advised when working with GetDayBar, as it could return null if there's not enough intraday data to build the virtual daily bar, so as you added the native daily series this was resolved - http://www.ninjatrader.com/support/h.../getdaybar.htm

[drolles]

Bertrand,

Thank for the prompt reply.

I was setting the Min Bars Required up to 80 – 100 (of 60 min bars) that is not enough to fill the indicator? That is nearly a weeks worth of data.

What would be in the min required?

Thanks and regards,

drolles

[NinjaTrader_Bertrand]

MinBarsRequired would mean how many bars for each series need to be present in order to call the OnBarUpdate() - so with this high setting you prevented the idea from working - how many days back of data were you loading?

[drolles]

Bertrand,

Thanks for your reply.

Your question was thought provoking so I’ve done some testing around this now.

I set the Min Bars Required to 20, again with 60 min bars as target timeframe with secondary timeframe of 1 min bars.

I’ve just got NT7 to crash backtesting my strategy.

Sequence of events:

1. Backtested on 2010 data (to date i.e. 22/10/2010) not connected to Esignal – backtestested fine
2. Restarted NT7
3. Backtested on 2010 data (to date i.e. 22/10/2010) not connected to Esignal – received Object exception message
4. Backtested on 2009 data not connected to Esignal – backtested fine
5. Connected to Esignal
6. Backtested on 2010 data connected to Esignal – NT7 said loading missing data items – crashed

Hope this helps with the beta.

Kind regards,

Dan

[NinjaTrader_JoshP]

Can you please provide us with your complete log/traces? Thank you.

Can you please provide us with the strategy you are using as well so we can see how exactly it is programmed? Thanks.

[drolles]

Josh,

No problem.

I have forwarded the log file referencing this thread. Unfortunately, I don’t think it is appropriate that I post the strategy here.

I hope it helps the beta development.

Kind regards,

Dan

[NinjaTrader_Austin]

Thank you, Dan.

[devdas]

@drolles

I had similar problem in one of Indicator coding.

Later i realised it depends on Instrument's Session template.
It was mismatching of Session template that causing Null return, that i could correct with correct session template.

Hope it helps.

[drolles]

Devdas,

Thanks for letting me know and the pointer towards the session template. I've not seen the problem for a while, but this is good for reference.

Kind regards,

drolles