Re: Winter Project

[Thomas Droege <tdroege2@earthlink.net>]

Michael,

That is the plan.  I have done this once before and it did not improve the
over all distribution but now that I understand that cause, perhaps it will
improve the individual star measurements.

The plan is to:

1) Group all the measurements of a star from all images together as at
present.
2) Look at the individual star measurement sets and copy the outlier (by
some definition) lines to a separate file.
3) Sort the outliers by measurement image.
4) Sort the whole data base by measurement image
5) Compute the ratio of outliers in each image as a percent
6) Add this outlier percentage to every measurement in each image by image.

Tom Droege

> [Original Message]
> From: Michael Koppelman <lolife@bitstream.net>
> To: <tdroege2@earthlink.net>
> Cc: tass <tass@listserv.wwa.com>
> Date: 10/31/2003 9:07:33 AM
> Subject: Re: Winter Project
>
> I don't know if this has been stated or not yet, but perhaps it's not 
> the stars we need to flag but the images? i.e. can we determine if a 
> given image is an outlier and then flag all the measurements from that 
> image?
>
> Michael
>
>
> On Oct 29, 2003, at 11:51 AM, Thomas Droege wrote:
>
> > The plan is to find all the outliers - say all those measurements for 
> > all
> > stars more than 3 sigma from the mean value for a star.  Then compute 
> > the
> > percentage of outliers in each frame.  Then flag all the readings with 
> > a
> > code as above.  Or I might just put a number in the flag position
> > indicating the percent "outliers" which would allow you all to make 
> > any cut
> > you like.
> >
> > Comments anyone?
>