How to give Upper limit on a branching ratio

Hi
I would like to calculate U.L. on the branching fraction. I am performing a simultaneous fit to the two sub decay modes. Background consists of a mixture of BF Gaussians and a threshold function and signal part consist of two Gaussian.

At present I am doing something like this:

  1. generating signal and background events.
  2. Fit the sample with simultaneous fit.
  3. calculate the events if they are not more than 90% of the central signal yield, I repeat step 1 by increasing the signal events.

This way I get U.L. (@90% C.L) using the frequentist method.

Now I want to give U.L. on the ratio of this branching fraction and the another well measured branching fraction. Here I am getting confused how to do this.

Any idea… or is there any other way to do this… I tried to look at Roostats but I didn’t get much… Any kind of tutorial or reference is also welcome…

Thank you