I have data of z values, distributed mostly uniformly. I want to fit it with the uniform distribution (I know that MLE will be just maximum and minimum values). How can I do that?
I tried RooFit::Uniform, but if I initialize my data with variable z with a given range, the very same range will be used when I plot this Uniform pdf. I know that I can select the range for plotting, but can I fit the range in question?
There are examples how to use variable limits for integrals, but I couldn’t find an example of the uniform distribution with variable limits.
I can copy my code if needed.
@jonas could you please take a look whenever you will have time? Thanks!
HI @ynikitenko! Variable range limits can be set in principle via RooRealVar::setRange, but I don’t think you can fit parameters if they are only used in for range limits.
It would be best if you can copy-paste your code such that we can discuss about the best solution for your problem. For example, I can imagine that is could be better to use a step function in your case.
Hi @jonas! Thanks for the reply.
I think that it is not possible to model a Uniform pdf with variable range in RooFit. I think I’ll make a feature request for that.
About the code - I already wrote my own pdf with other parameters. I don’t know what could be substituted for the uniform pdf. It should just exist, IMHO.
What step function do you mean? The Heaviside theta function?
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.