From: Robert A. Knop Jr. (robert.a.knop@vanderbilt.edu)
Date: Thu May 01 2003 - 04:33:28 PDT
On Wed, Apr 30, 2003 at 08:22:45PM -0700, Greg Aldering wrote:
> I understand that it is hard to code for all the different types of
> observational situations which may arise, but to remove from the Hubble
> diagram a SN based on such hard-won and expensive data, without first
> imposing every realistic constraint on the spectral fits, just doesn't
> make sense to me.
This message is mainly for Andy and Peter, but I want to make sure Greg
sees it since he's active in this discussion too.
Several messages back I asked this question, but I think it got lost in
all the back-n-forth.
Way back when Andy sent me a list of "Definative Ias", "Probable Ias",
and "Dunnos". The primary subsets (used for most of the analysis in the
paper) omit the "Dunnos", but keeps the Probables. Subset 3 (used only
in the systematic error section for "type contamination") omits the
Probables.
My question is: are the 9878 and 98104 spectra consistent with the
Probable Ias? If so, then all we have to do is remove them from Subset
3, and the only section which can change at all is that systematic error
section. We leave them on the Hubble diagram for the rest of the paper,
and nothing else in the analysis changes. This is a low pain way out,
and indeed I would prefer this way out to arguing for another week or
two about whether or not they should be in the primary subset or
nowhere. (That is, if in fact this is honest it does just make sense to
put them in Subset 3 given what we know today.)
Even if this is a "too conservative" approach, and we later decide they
should have been in the primary subset, for purposes of estimating a
systematic error that's not such a terrible thing.
-Rob
-- --Prof. Robert Knop Department of Physics & Astronomy, Vanderbilt University robert.a.knop@vanderbilt.edu
This archive was generated by hypermail 2.1.4 : Thu May 01 2003 - 04:33:32 PDT