It turns out to be possible to use the "Customize" and "Edit" features
of graphs to do this although the end result may not have all the
properties you might want in a graph. The method is both elegant and
clunky at the same time! Basically you create two completely separate
charts of the same type one for Y1 and one for Y2. You use Customize to
make one or both of them look the way you want. Then you copy the frame
contents of the graph of Y2, say, and paste the frame contents onto the
graph of Y1. The graph you copy is not live to the data in the resulting
plot. Also if you save the script for the combined plot the pasted part
is ignored. You can, however, journal the result for later use. You can
probably script the copy/paste but I haven't tried it. Using this
technique I was able to more or less replicate Mark's graph without
scripting.
Michael
----- Original message -----
From: "Mark Bailey" <mark4flies@aol.com>
To: jmp-l@lists.biostat.wustl.edu
Date: Tue, 28 Jul 2009 20:29:37 -0400
Subject: Re: [jmp-l] Add markers for model-predicted values to a
variability chart
I think that you will need a script to make such a chart. It would not
be too difficult.
How are you using the Variability Chart to analyze data from a DOE?
On Jul 28, 2009, at 4:40 PM, Bruey, Francis wrote:
> Is there a way to add markers for another column of values (e.g.
> values from a prediction formula saved to a table from Fit Model) to
> an existing Variability Chart (e.g. a chart of raw experimental
> response values)?
>
> Or do I need to write a script to do this?
Mark
They hate you if you're clever, and they despise a fool ("Working
Class Hero" - John Lennon)
|