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: censored regression - testing for normal and homoscedastic residuals
From
Maarten Buis <[email protected]>
To
[email protected]
Subject
Re: st: censored regression - testing for normal and homoscedastic residuals
Date
Tue, 31 May 2011 17:59:03 +0200
On Tue, May 31, 2011 at 5:37 PM, Anders, Christoph
<[email protected]> wrote:
> I conducted a regression with right-censored variables using the cnreg command.
> Since I suppose the residuals to be both heteroscedastic and not normally distributed, I would like to conduct some tests to proof it.
> But the commands I usually use - estat hettest and estat imtest - seem not to work here.
>
> Anybody who knows how to test for normally distributed and heteroscedastic residuals when having used cnreg?
You could sample datasets from your model, estimate a new model in
each of these datasets and store the residuals. These will give you an
idea of how the distribution of residuals should look like when the
model is correct. Graphically comparing these with the actual
residuals (e.g. using Nick Cox's -qplot- which you can find by typing
in Stata -findit qplot-) would be a good way of making your suspicions
plausible.
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/