Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: replicability of bootstrap standard errors in quantile regression


From   "Mak, Timothy" <[email protected]>
To   <[email protected]>
Subject   st: RE: replicability of bootstrap standard errors in quantile regression
Date   Tue, 20 Mar 2007 14:18:07 -0000

This seems normal. If you set your seed now, and run your do file, after
you finish your do file, the seed is changed, and so if you run your do
file again, you get different results. If you set your seed within your
do file, however, every time the do file is run, the seed is set to the
same number and hence you get the same result. 

Tim  

-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Jessica
Goldberg
Sent: 20 March 2007 03:17
To: [email protected]
Subject: st: replicability of bootstrap standard errors in quantile
regression

I am having trouble reproducing bootstrap standard errors, despite
setting the seed immediately before the regression command.  If I run
the same .do file twice, I get different standard errors each time.

However, if I repeat the code within the same .do file (re setting the
seed), I get the same standard errors within the file.

The lines of code I am using are:

set seed 123456789
bsqreg earnwke sex age age2 ed_9to10-ed_16ormore, q(.5) reps(50)

I've had the same results setting the seed in the same line as running
the regression.

I'd appreciate any suggestions.  Thank you for your help.

Jessica Goldberg
[email protected]
PhD student, Department of Economics
University of Michigan

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index