Using RooUnBlindPrecision

Hello RooFit experts,

I am trying to do a fit by blinding a fit parameter (tau see code snippet)
Here is what I want to achieve:

  • Say the parameter to blind is tau, I want the Roofit to use tau for likelihood construction
  • But the result printed by Roofit should have blinded parameter tau_blind which is randomly shifted from real tau by ± 5*sigma (sigma is some number)

I tried using RooUnblindPrecision following several posts in this forum
Below is the relevant code snippet I have used:

RooRealVar tau("tau","tau",0.000104,0,0.001);
RooUnblindPrecision tau_blind("tau_blind","tau_blind","taublinded",0.000504,0.000004,tau,0);
RooDecay decay("decay","lifetime",t,tau_blind,gauss1,RooDecay::SingleSided); 

According to available documentation, the arguments are:

RooUnblindPrecision name(“name”,“title”,“blindstring”,central value,scale,blindvalue,sin2betaMode);

But I don’t understand the meaning of these parameters
central value: ?
scale: ?
blind value: parameter to be blinded
sin2betaMode: ?

Can anyone please explain?
Or any other way of implementing what I am trying to do here in Roofit
other related function I came across RooUnblindUniform

Related previous posts:

Hi @amroo - I have invited @moneta and @jonas to help, but due to the Christmas break this might take a bit longer than we’d hope.

1 Like

Unfortunately we are missing a tutorial in Roofit for using these unblind tools.
I think the RooUnblindPrecision applies a simple transformation to the parameter mantaining the same fit precision. It is simply applying a translation defined by the scale around a given central value. See the corresponding code in



This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.