Re: Fits/Kcorrs analysis for P99 complete

From: Greg Aldering (aldering@panisse.lbl.gov)
Date: Fri Mar 21 2003 - 23:19:10 PST

  • Next message: Robert A. Knop Jr.: "Fun with chisquares (more fits put on HST paper web page)"

    Hi Rob,

    Thank you for calculating and displaying the changes to the P99 SNe when
    different treatments are used. This has been very helpful to me in
    understand how the fits migrate as didn't assumptions/corrections are
    made. I do have a few questions about those comparisons that you have
    already done which I hope you can answer.

    1) What is the sign of the "Differences" tables? Is it "filled" minus
       "stroked"?

    2) For A-2, did the stroked fit use your new lightcurve template?
       If so, would you mind adding that to the specification?

    3) In the "Differences" table is the "chirat" column the ratio of the
       SNminuit lightcurve fit Chi-squares?

    4) Would it be possible to add to the differences tables the deviations
       from the best-fit cosmology, in units of magnitudes and
       error-normalized, as well as the differences in these? This would help
       which SN are the biggest outliers and in determining why the
       cosmology fit Chi-squared values get progressively worse as you add
       more changes relative to P99.

    I do note that between the P99 fit of line A-1, and the full treatment
    you would like to use on A-5, the *change* in Chi-squared is 8.2. Since
    the *change* in Chi-squared is a useful means of hypothesis testing, we
    see that the final A-5 fits are almost 3-sigma worse that the original P99
    fits. (I grant that P99 had a MW extinction error, but I wasn't sure if
    you have a fit where that was the *only* fix you made - see question 2
    above - so I am comparing to the original P99 fits to be concrete.) This
    is why I would like to see whether the deviations for all the SNe got
    worse, or whether there are certain SNe which became much worse (my
    request 4, above).

    I'm sorry for the extra requests, but for me these comparisons have been
    very valuable, and with a little more information I think we (or the
    rest of us) will understand much better the foundations for the HST paper.

    Thanks!

    Greg

    This archive was generated by hypermail 2.1.4 : Fri Mar 21 2003 - 23:19:12 PST