| Date: | Sun, 25 Sep 2005 21:24:07 -0700 |
|---|
| Reply-To: | David L Cassell <davidlcassell@MSN.COM> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | David L Cassell <davidlcassell@MSN.COM> |
|---|
| Subject: | Re: MDC, nested logit, weights |
|---|
| In-Reply-To: | <200509211725.j8LH4pHh016225@malibu.cc.uga.edu> |
|---|
| Content-Type: | text/plain; format=flowed |
|---|
polyama@YAHOO.COM wrote:
>It seems that proc MDS does not have weighing through WEIGHT or FREQ.
>How can I estimate nested logit model when observations need use of
>weights?
Well, that depends on *why* you have weights.
If the weights are not sampling weights from a sampling design, you
can build a nested logit model using PROC NLMIXED in the SAS/STAT
package. You can include weighting there.
If the weights *are* sampling weights from a sampling design, then
you've got a problem. I would try to do what I could with PROC
SURVEYLOGISTIC, but that is certainly not designed to handle a
nested logit model. How many levels of nesting are you building?
HTH,
David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330
_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
| 
