I am not very certain if you can do it, if you do not have any guess coefficients. It is a known problem that if the guess coefficients are too far away from the real values, the fitting can go wild.
However if you have some sort of idea what the coefficients-area, offset,FWHM and center are, you can just use a normal Lev-Mar fitting function to do the job. -Fan "pshuk256" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] > I want to fit my data to some gaussian. I don't know any coefficients. > Can anyone tell me how can I compare my data set to some arbitrary > gaussian.? > > In anticipation of reply. > priiyank
