Hello evrybody i’m new root user i had a lot of problem to use Miinuit so when i want to minimize the chi square function a had all the time a high correlation between parameters is there a way to reduce these correlations ??with reparametrization of the function it dosen’t work i had always the same parameter correlation cofficients
thank you very much
Minuit cannot do anything to reduce the parameter correlation. It is up to the user to re-define the parameters in a way that is more appropriate. For example if two parameters are 100% correlated you can fix one and just use the other one in the fit.
You could use a linear transformation of the parameter which diagonalise the covariance matrix