Your question lacks enough details to be certain about what you are doing but if you are fitting a Gaussian distribution defined by exp(-(x-m)^2/(2 s^2))/z then the result can't be negative no matter what.
It is possible that your calculation of the gaussian is buggy. Or it is possible that your plotting is broken. That can easily happen if you are using splines to smooth your plot. Sent from my iPhone On May 13, 2012, at 12:06 AM, "B.Kumar" <[email protected]> wrote: > Hi! > I am using the gaussian fitter. Occasionally, I see that the gaussian curve > that it generates goes negative - undershoots the x-axis. How do I avoid this? > Please see the attached plot (using JFreeChart). > Thanks in advance for your help. > B.Kumar > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
