Re: Saul's first suggestion (fwd)

From: Don Groom (deg@panisse.lbl.gov)
Date: Tue Mar 18 2003 - 11:08:39 PST

  • Next message: Robert A. Knop Jr.: "procedure last time"

    ---------- Forwarded message ----------
    Date: Tue, 18 Mar 2003 12:46:22 -0600
    From: Robert A. Knop Jr. <robert.a.knop@vanderbilt.edu>
    To: Don Groom <deg@panisse.lbl.gov>
    Subject: Re: Saul's first suggestion

    On Tue, Mar 18, 2003 at 10:31:13AM -0800, Don Groom wrote:
    > Rob,
    >
    > Here's more or less Saul's first suggestion for StupidPrior.
    >
    > In the 2nd, the columns and rows are interchanged, and a 4th case is added
    > to the rows--only one of these cases (besides hi-z) is not (yet)
    > available, and I can't actually remember what it is. Won't have it ready,
    > as Saul has become virtual.

    A few comments:

    First, don't knock yourself out, since *all* of these fits are going to
    be redone.

    Second, I look at this and suffer information overload. Too many
    panels, it doesn't get a simple point across.

    Third, I don't want to show fits with just high-z team data, for two
    reasons. One, it's not their paper. Two, we will have no end of hell
    when people ask why our fits to their data give different confidence
    intervals than their fits. (There are reasons, of course, but I really
    just don't want to go there.)

    I'm planning instead on a figure with just two or three panels,
    comparing uncorrected and currected with the *full* set with and without
    the prior.

    -Rob

    -- 
    --Prof. Robert Knop
      Department of Physics & Astronomy, Vanderbilt University
      robert.a.knop@vanderbilt.edu
    


    This archive was generated by hypermail 2.1.4 : Tue Mar 18 2003 - 12:05:17 PST