[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: updated talk for PID



Hi Eugene,

    I did not look at the difference in energy resolutions because it is a bit more complicated than that. It is not just delta(E) that is constrained to come out to be 0, but the missing mass as well. For a 4-C fit, just looking at delta(E) and the subsequent resolution is ignoring other information in the fit. Also, after the fit, delta(E) is a spike at 0 by design, regardless of what mass hypothesis I use. Before the fit, I have no doubt this distribution will look exactly like yours as we are using essentially the same resolution functions.

   For the kinematic fit, the best measure of the goodness of fit is the confidence level, which is flat for the proper distribution and lets us know that the fit is working the way it is supposed to.

    The best measure of the goodness of bkgd/signal separation is a bit more difficult to say, as it depends on your PID scheme and how much signal you are willing to give up (what anticuts are made). But to first order, as a visual guide, I would say the slides that show all the CL distributions have much fewer events above 10% demonstrate that it is a challenge to get the 4pi background to look like a 2pi2K event.

    For a qualitative statement of how well we separate that particular signal from that particular background, the numbers Curtis had are the right statement to make. I'm not saying that we should add those to any talk....but they are what they are.

Matt


Reading the 999 paper and your presentations I realized
that the difference is significant and starts, perhaps,  at the initial
resolutions. However, the track resolutions I use (taken from David) are
similar or only slightly worse than those you show on page 5 of your
presentation.
Let us consider the reactions you used:
1) gp->pK+pi+K-pi-
2) gp->ppi+pi+pi-pi-

These reactions are identical in the 3-momentum space. The only
difference is the energy. It is a 1-dim. separation.
I obtained the E_final-E_initial resolution of sigma=150 MeV
(180 MeV for some PYTHIA events, perhaps because of a different energy
distribution). The reaction 2), treated as 1) gives an average energy shift
of 120MeV, which is 0.8 sigma of one peak. So, I get a rather small separation...
 What are your values for these resolutions/separations?

Thanks,
Eugene

------------------------------------------------------
Eugene Chudakov
http://www.jlab.org/~gen
phone (757) 269 6959  fax (757) 269 5703
Thomas Jefferson National Accelerator Facility
12000 Jefferson Ave, Newport News, VA 23606 USA


--
--
----------------------------
Matt Bellis
Carnegie Mellon University
(office) 412-268-6949
(cell) 412-310-4586
----------------------------