| Date: | Sun, 11 Aug 2002 05:40:27 +0200 |
|---|
| Reply-To: | Getz <getz@gadot.org.il> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU> |
|---|
| From: | Getz <getz@gadot.org.il> |
|---|
| Subject: | Relative importance of variables in multinomial regression |
|---|
| Content-Type: | text/plain; charset="us-ascii" |
|---|
Hi list,
Is there a way to identify the effects of different IV in multinomial
regression?
Menard (1995) and Long (1997) proposed how to calculate the standardized
deviation, and it can be done also for every category of the DV in
multinomial regression, but I want to estimate the relative importance
of the IV for the complete model. Can I use The likelihood ratio test as
a measure of the order of importance of my variables? Any other
suggestion?
TIA
Shlomo Getz
The Institute for the Research of the Kibbutz
University of Haifa
Tel:
Home: 04-6939150
Work: 04-8240418
|
