Event by event resolution function modeling with Dt_{err} as a conditional observable

[sagar_hazra]

I want build an event by event resolution function where Dt_{err} will be conditional observable. I am attaching here a code including the result. Basically I am not sure about the result, without giving the conditional argument in the FitTo(data) function I am getting the same result. Please let me know where I am wrong.
Resolution_function_perevent_post.C (4.1 KB)
result.txt (855 Bytes)

[jalopezg]

Hi @sagar_hazra,

I cannot see anything obviously wrong in the attached macro, so I will let our RooFit experts @jonas and @moneta take a look.

Cheers,
J.

[sagar_hazra]

@jalopezg , many thanks for looking at the code and provide an important feedback.

[jonas]

Hi @sagar_hazra,

you are doing everything right! Indeed, it is expected that the result doesn’t change if you don’t specify the ConditionalObservables() in fitTo.

One thing you are still missing is to plot the model correctly together with the data. When plotting your model, you should to the projection of dterr based on the dataset, as explained in the documentation for RooAbsReal::plotOn:

resolution.plotOn(res_frame, ProjWData(*data));

Then, also the plot will make sense to you I hope!