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, statalist.org is already up and running.


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

RE: st: bootstrap _b VS bootstrap _se


From   "Lachenbruch, Peter" <Peter.Lachenbruch@oregonstate.edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: bootstrap _b VS bootstrap _se
Date   Wed, 21 Jul 2010 09:02:30 -0700

The grumpy old man in me also suggests that when on the boundary of 'significance' you need to be very careful as we are usually in the multiple testing scenario and we don't want to overstate our enthusiasm.  I'd report the results, but downplay such statements.

Tony

________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten buis [maartenbuis@yahoo.co.uk]
Sent: Wednesday, July 21, 2010 1:26 AM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: bootstrap _b VS bootstrap _se

--- On Wed, 21/7/10, Sirak wrote:
> But what if the coefficients which I found them significant
> ( with high t-values) in SUREG are not more significant in
> bootstrap _b, reps(1000). Isnt it this means the orginal
> findings dont replicate??

No, these are two different ways of estimating the standard
errors. There exist (often conflicting) arguments you can use
to choose one method over the other, but if you find that the
method matters, then I would just conclude that we don't know
whether that statistic is significant or not. Some people use
the term "on the boundary of significance".

Remember that the standard error, and thus the p-values and
confidence intervals, are also just estimates, and thus are
themselves also uncertain. So, it should not come as a
surprise that there exists a grey area where we don't know
whether a statistic is significant or not. Normally we don't
estimate the uncertainty in our standard errors, so we don't
notice it. We only tend to find this grey area when we
compare different methods of estimating the standard errors,
as you found out.

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/
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index