I’m trying to figure out the best way of fitting a background and signal function to a histogram.
In the “good ol’ days” of mn_fit, I’d:
- exclude the signal region
- fit the background function
- fix the background function parameters
- fit the signal region with the signal + background function
- release the background parameters and fit everywhere
There seem to be two impedements in Root to doing this:
- releasing parameters
- excluding regions
As for the former, I can use ‘SetParamLimits’ to release the parameter, but is there an easy way to do this from my TF1 without enforcing a range?
As for the latter, you can write a special function that ignores certain regions, but is there a way you can use the built in functions and just tell the fitting to ignore (temporarily) a region?
Thanks,
Charles
p.s. If you currently can’t do either of what I asked, would it be possible to put it on the list for future upgrades?