[prev in list] [next in list] [prev in thread] [next in thread]
List: sas-l
Subject: Re: Odds ratio table for cumulative logit model
From: "Gravelle, Tim" <Tim_Gravelle () GALLUP ! COM>
Date: 2011-05-30 13:25:26
Message-ID: 64C85429AE96F945BCA136EAC7588324FF4742C3 () exchng7 ! noam ! gallup ! com
[Download RAW message or body]
I would suggest that interaction effects require special (or at least especially \
careful) interpretation in all models, linear, logit or otherwise. :) I can \
recommend the following sources covering the specification, interpretation and \
display of interaction effects (including their use in logit models):
Braumoeller, Bear F. (2004) "Hypothesis Testing and Multiplicative Interaction \
Terms." International Organization 58(4): 807-820.
Fox, John and Robert Andersen (2006) "Effect Displays for Multinomial and \
Proportional-Odds Logit Models." Sociological Methodology 36: 225-256.
Jaccard, James (2001) Interaction Effects in Logistic Regression. Thousand Oaks, CA: \
Sage.
Tim
-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of \
polyphonic48@gmail.com
Sent: Sunday, May 29, 2011 3:18 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Odds ratio table for cumulative logit model
I have run a cumulative logit model using proc logit and have found that the odds \
ratios for two (continuous) variables used in an interaction, and the ratio for the \
interaction itself, did not print. I recall reading somewhere that interactions \
require special interpretation in logit models but the texts on my shelf do not cover \
interactions.
Any advice, recommended reading, etc. would be appreciated! Thanks.
IMPORTANT NOTICE: This e-mail message and all attachments, if any, may contain \
confidential and privileged material and are intended only for the person or entity \
to which the message is addressed. If you are not an intended recipient, you are \
hereby notified that any use, dissemination, distribution, disclosure, or copying of \
this information is unauthorized and strictly prohibited. If you have received this \
communication in error, please contact the sender immediately by reply e-mail, and \
destroy all copies of the original message.
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic