[G12] crystal ball function

[Craig Bookwalter]

Hi MK,
  My intuition is not well-developed in this arena, but in your binned fits
of the pi0 yield, it seems to me that a reverse Crystal Ball function would
better suit your data, given the tails that stick out on the high side.
This shape is included with RooFit (which I have never used myself but
seems to be heavily used in HEP) as RooCBShape. I found a tutorial online
for RooFit that includes a CB lineshape (written by a guy I went to grad
school with, actually):

http://hadron.physics.fsu.edu/~skpark/document/ROOT/jdhaas_RooFit.pdf

However he uses a regular CB, whereas I think you need a reverse CB. From
the following link, I think you can get a reverse CB from requiring the
parameter "n" to be negative:

http://root.cern.ch/phpBB3/viewtopic.php?t=6196

There is also a Wikipedia article about it:
http://en.wikipedia.org/wiki/Crystal_Ball_function

I am sorry that I don't have any direct experience with which to help you,
but perhaps this can get you started. Also, I checked my flights, and I'll
be able to come to the first 20 minutes of next week's meeting.

Good luck!

--cb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailman.jlab.org/pipermail/g12/attachments/20131210/721ff1f9/attachment.html