Error on the yield

Hi

One simple question: when I use the maximum likelihood with (for instance) different functions for signal and background, how the error on the signal is obtained ?

Regards

Bruno

HI,

I am not sure I have understood well your question. In general, by doing a global fit to the signal and the background you will get the error on the signal.
Regards,
Lorenzo

My question is how is calculated the error when you use the maximum likelihood method ?

The error can be obtained from the fit, from the inverse of the Hessian matrix (covariance matrix) in the approximation that the likelihood is a parabola around the minimum or from a profile likelihood scan
(Minos method)

See seal.web.cern.ch/seal/documents/ … nerror.pdf

Lorenzo