Date:         Fri, 25 Jul 2008 16:28:02 -0700
Reply-To:     Jeff <jeffrey.m.allard@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Jeff <jeffrey.m.allard@GMAIL.COM>
Organization: http://groups.google.com
Subject:      Re: Propensity scores
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset=ISO-8859-1

On Jul 25, 11:27 am, ohri2...@GMAIL.COM (ajay ohri) wrote:
> Hi Peter,
>
> then what is the alternative to say creating propensity scores on leads
> through direct marketing channel (direct mail,internet,call centre)..
>
> regards,
>
> Ajaywww.decisionstats.com/tag/analytics
>
> On Fri, Jul 25, 2008 at 8:47 PM, Peter Flom <
>
> peterflomconsult...@mindspring.com> wrote:
> > The topic of propensity scores has come up a couple times recently.
>
> > People interested in these should check out the latest American
> > Statistician, especially:
>
> > Peikes, DN; Moreno, L; Orzol, SM (2008).  Propensity score matching: A note
> > of caution for evaluators of social programs.  American Statistician, 62,
> > 222-231.
>
> > Essentially, they found that, even under ideal conditions, propensity
> > scores give incorrect results.
>
> > Peter
>
> > Peter L. Flom, PhD
> > Statistical Consultant
> > www DOT peterflom DOT com

Ajay-

In what sense? Certainly randomization and a designed experiment is
the way to go when testing or modeling in the DM field - which is
where I work.

Jeff