--- On Sat, 11/7/09, Murtaza, Fowad wrote:
> I am bit confused understanding the stata help comment on
> sqreg. Does it mean that we can directly compare the
> confidence intervals if we want to confirm that coefficient
> of say quantile 10 is different from coefficient of
> quantile 20?
No, but that has nothing to do with -sqreg-. If you want to
test whether two estimates (whatever estimates) are equal
you cannot do so by looking if the confidence intervals
overlap.
> In fact, after estimating sqreg for 10 quantiles, I want to
> see if the coefficient of a variable (e.g. urban) in
> quantile 10 is significantly different coefficient of same
> variable (urban) in quantile 20 or 30?
You can use the -test- command, like in the example below.
*------------ begin example -------------
sysuse auto, clear
sqreg price weight length foreign, ///
quantile(.25 .5 .75) reps(100)
test [q25]weight = [q50]weight
*------------ end example ---------------
Referring to your question you would type something like:
test [q10]urban = [q20]urban = [q30]urban
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
