Hi All,
I am having a bit of trouble with RooFFTConvPdf. When I try to convolve a test function with a Gaussian about the origin, I am seeing a shift in the resulting convolved function that I don’t understand. I have written a test script that demonstrates this using a couple of convolved Gaussians and the difference between the numeric treatment and the analytic treatment. This feature does not show up in RooNumConvPdf, but I am reluctant to use that, as there are warning all over its documentation.
Please let me know if you have any suggestions.
Cheers,
Luke
fit_work.cpp (1.29 KB)