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]

st: RE: Bootstrapping to get Standard Errors for Regression Discontinuity Estimators


From   Nick Cox <n.j.cox@durham.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Bootstrapping to get Standard Errors for Regression Discontinuity Estimators
Date   Thu, 23 Sep 2010 12:12:20 +0100

Jen Zhen: 

You sent this email earlier on 20 September and it got no replies. In principle, I see four reasons for your perceived problem. 

1. The theory underlying -rd- you are using is wrong, in that there are incorrect deductions from its assumptions. 

2. The theory underlying -rd- does not fully apply to your data because assumptions are not satisfied in some serious sense. 

3. Austin's code or Stata's bootstrap code is incorrect. 

4. Your intuition is at fault. 

Naturally, the explanations are not mutually exclusive. 

My own guess is that your question is unanswerable. An expert in the field would doubt 1, anyone experienced in Stata would doubt 3, and we have no way of commenting on 2 or 4. 

As the Statalist FAQ advises, it is better to revise a question not answered than just re-post it. 

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

Jen Zhen 

When bootstrapping Austin Nichols' rd command:

bs, reps(100): rd outcome assignment, mbw(100) ,

I find that often the resulting P value tells me the estimate is not
statistically significant at the conventional levels, even when visual
inspection and more basic methods like simple OLS regressions on a
treatment dummy, assignment and assignment squared suggest huge
statistical significance.

That makes me wonder whether possibly this boot-strapping method might
somehow understate the true statistical significance of the effect in
question? Or can and should I fully trust these results and conclude
that the estimate is not statistically significant at the conventional
levels?


*
*   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