Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Constant term in Logit regression


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Constant term in Logit regression
Date   Mon, 22 Apr 2013 12:07:09 +0200

On Sun, Apr 21, 2013 at 1:15 PM, Athinagoras Konstantinidis wrote:
> However, the constant term has  a p-value of about 0.600.

The p-value of the constant term in logistic regression belongs to a
test with a rather peculiar null hypothesis, see the final comment of
(Buis 2012). So before you are going to draw conclusions based on that
p-value you first need to understand that null-hypothesis and decide
wether or not you care about that. In all likelihood you don't care
about that null-hypothesis and the p-value is thus meaningless.

On Sun, Apr 21, 2013 at 10:05 PM, JVerkuilen wrote:
> You hardly ever interpret the constant unless it's been scaled to be
> interpretable.

This is true but the consequence is that you should scale your
variable such that the constant is interpretable and _not_ that you
should never interpret the constant. Especially in a logit model,
correctly interpreting the constant can be extremely helpful, see e.g.
(Buis 2012)

Hope this helps,
Maarten

 M.L. Buis (2012) "Stata tip 107: The baseline is now reported", The
Stata Journal, 12(1), pp. 165-166.

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index