st: RE: RE: Problem with grqreg

["Martin Weiss" <martin.weiss1@gmx.de>]

<>


Not so sure: This code comes back w/o any panel for "foreign" and "trunk":

***********
webuse auto, clear
sqreg price weight length turn foreign /* 
*/ trunk, q(.1 .2 .3 .4 .5 .6 .7 .8 .9) /* 
*/ reps(400)
grqreg, cons ci ols olsci scale(1.1)
***********


HTH
Martin

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: Mittwoch, 5. Mai 2010 19:47
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: Problem with grqreg

I tried this without problems:

clear
set obs 100
foreach v in y x1 x2 x3 x4 x5 x6 { 
	gen `v' = rnormal()
}
sqreg y x1-x6, quantile(.10 .20 .30 .40 .50 .60 .70 .80 .90) reps(400)

grqreg, cons ci ols olsci scale(1.1)

So, my only guess is that this is something to do with your data. 

Nick 
n.j.cox@durham.ac.uk 

Michael C. Morrison

First I run sqreg  with the following commands:

set seed 10101
sqreg dep_var  iv1 iv2  iv3 iv4 iv5 iv6, quantile(.10 .20 .30 .40 .50 
.60 .70 .80 .90) reps(400)

I want to use grqreg to plot the changes in the quantile slopes. I use 
the following command:

grqreg, cons ci ols olsci scale(1.1)

which should, I think, plot all 6 independent variables, plus intercept 
changes, with OLS and its ci added to each plot.

The problem is that the command only prints the intercept and the first 
two covariates (with OLS and its ci).

Thinking this was a limitation of available graph space I then tried to 
specify the next three covariates in the [varlist]:

grqreg  iv3 iv4 iv5, cons ci ols olsci scale(1.1)

Unfortunately I receive the same exact outcome as in the first grqreg 
command.


*
*   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/

*
*   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/