Finished with 42 refits

From: Alex Conley (aconley@panisse.lbl.gov)
Date: Fri Mar 14 2003 - 16:13:23 PST

  • Next message: Robert A. Knop Jr.: "Re: comparison of mb and stretch, old and new"

    I've finished checking the 42 SNe. My hands/wrists are hurting, so I'm
    going to take off. Somebody else can deal with the HST SNe.
    This time I'm not going to attach all of the files because they are
    rather large (Okay, it was big last time too, I just didn't notice).
    Instead, I'll tell you where to look for them:

    /home/astro51/u/aconley/hst/robrefits/

    This time I've only attached a comments file going through each SNe
    individually. Basically, there are some crappy ones in the 42 that
    have fits which are unstable to the initial conditions. Moreover,
    they are unstable in a fashion that leads to two very different fits
    with about the same chi-square. Feel free to take a look at the
    plots for yourself -- they are in the snminuitoutput.42.tgz file.
    As before, there is a gnumeric spreadsheet which summarizes the
    differences between fits.

    Examples of this are sn921, 94201, sn9794. These should probably be
    excluded from all fits on these grounds.

    I also checked fixed vs. floating where I thought it appropriate.
    There seems to be some mismatch between my list and Rob's. For example,
    I think that 9785 should have been fit with floating V.

    This exercise also confirmed what I already knew about the error bars --
    the MINOS ones are much more trustworthy. It's too bad there not the
    ones that go into the error matrix (I think).

    Maybe I'll come in tomorrow after the peace protest and work on the HST
    ones, but maybe not.

    Alex





    This archive was generated by hypermail 2.1.4 : Fri Mar 14 2003 - 16:13:24 PST