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: st: positive interaction - negative covariance
From
"JVerkuilen (Gmail)" <[email protected]>
To
[email protected]
Subject
Re: st: positive interaction - negative covariance
Date
Fri, 22 Feb 2013 13:10:27 -0500
On Fri, Feb 22, 2013 at 12:32 PM, andrea pedrazzani
<[email protected]> wrote:
> Also, both b1 and the combined coefficient (b1+b3) are positive, but
> the covariance between b1 and b3 is negative. It sounds strange to
> me...
This kind of thing is all too common in practice with interactions,
which are often soaking up some often unimportant nonlinearity. This
can be created by the nature of the dependent variable, if there is
ceiling or floor effect, for instance, or a bit of skew. I concur with
Nick that graphing will be very helpful.
*
* 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/