Date:         Sun, 30 Mar 2008 12:34:00 -0500
Reply-To:     Wensui Liu <liuwensui@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Wensui Liu <liuwensui@GMAIL.COM>
Subject:      Re: Heckman correction in Proc Mixed?
Comments: To: Matt <mshall2@gmail.com>
In-Reply-To:  <eaf1faf5-1c03-4b08-b119-354a1dfb67c4@s50g2000hsb.googlegroups.com>
Content-Type: text/plain; charset=ISO-8859-1

hi, matt,
the key of heckit model is to get the inverse mills ratio? could you
do so using a probit model, calculate the IMR, and then plug it into a
mixed model?

On Sat, Mar 29, 2008 at 10:34 PM, Matt <mshall2@gmail.com> wrote:
> Hello,
>
>  I am wondering if anyone knows if it is possible to run a heckman-
>  style selection correction model using proc mixed? Or, are there other
>  techniques for running heckman models of multilevel data?
>
>  Thanks,
>  Matt
>



--
===============================
WenSui Liu
ChoicePoint Precision Marketing
Phone: 678-893-9457
Email : wensui.liu@choicepoint.com
Blog : statcompute.spaces.live.com
===============================