Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: AW: st: AW: Fitted probabilities using prvalue for logit model
From
Maarten buis <[email protected]>
To
[email protected]
Subject
Re: AW: st: AW: Fitted probabilities using prvalue for logit model
Date
Fri, 16 Jul 2010 14:02:56 +0000 (GMT)
--- On Fri, 16/7/10, Marc Michelsen wrote:
> Does your statement mean that -prvalue- is not an
> appropriate measure to determine the relative
> importance of my additional dummy variables relative
> to the benchmark model with its explanatory variables?
Finding out how much a variable adds to a non-linear
model is a surprisingly complicated issue. The
probabilities will tend to be closer to 50% in the
model without the added variables nett of what you would
get from the normal confounding or intervening
mechanisms. See for example: M.L. Buis (2010) "Direct
and indirect effects in a logit model", The Stata
Journal, 10(1), pp. 11-29.
Hope this helps,
Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/