Bookmark and Share

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]

st: Yearly interaction terms in Oprobit and multicollinearity


From   micdepo mic <micdepo@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Yearly interaction terms in Oprobit and multicollinearity
Date   Tue, 29 May 2012 22:26:38 +0300

Hi,

I  am trying to find the changing effects of a variable X on Y. Please
note that both X and Y are actually Rating variables. Thus, I am
running an ordered probit model on Y.  I also control for other
variables that may effect Y independently of X.

I have dummies for each year, and I also interact these year dummies
with X in order to find how each year X effected Y. I am looking for
an increasing or decreasing pattern in how  X effects Y. However, in
my postestimation analysis I find that VIF of X and the year dummies
and their interactions are as high as 30 to 40. I would very much
appreciate if you could let me know what your opinion on how I should
tackle issue. Maybe centering? I would also love to hear any opinions
on another potential model specification. Please remember I want to
learn the changing influence of X on Y.

I would appreciate any level of help or commentary. Thank you very
much in advance.

Best regards,
M
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index