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 on April 23, and its replacement, is already up and running.

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

Re: st: xi function dummies being dropped due to collinearity

From   "JVerkuilen (Gmail)" <>
Subject   Re: st: xi function dummies being dropped due to collinearity
Date   Sat, 25 May 2013 20:57:20 -0400

On Sat, May 25, 2013 at 3:34 PM, Lisa Marie Yarnell
<> wrote:
> Hello Statalisters:  I'm using the xi function to create school
> dummies to control for school effects in a quantile regression model,
> using the code below.  However, the dummies are all being dropped
> because of collinearity.  Why would this happen, and how can I
> alleviate this?  Thank you, Lisa

It must be the case that the school IDs are collinear with other
predictors in your model. I would suspect that the N in the dropped
schools is fairly small and that your other predictors end up being
collinear with the dummy, but that's just a guess. Stata might have
done you a favor because I wouldn't be surprised if the dummies end up
being nearly but not totally saturated in other schools, which would
not be happy.

More broadly I don't know what the current literature on quantile
regression suggests in the case of clustering of observations, so I
hope someone else chimes in. Is there a random effects model? (A bit
of Google shows:

JVVerkuilen, PhD

"They were careless people, Tom and Daisy - they smashed up things and
creatures and then retreated back into their money of their vast
carelessness, or whatever it was that kept them together, and let
other people clean up the mess they had made." -- F. Scott Fitzgerald
*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index