Re: st: Bootstrapping Error

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Thu, 7/10/10, Lina Marcela Cardona Sosa wrote:
> I am using bootstrap in STATA 11 to estimate the standard
> errors of a bivariate probit constrained to some restriction.
> I have written a program to estimate the bivariate probit
> subject to this particular constrained and then I use bootstrap
> over this program :
> bootstrap _b _se, reps(100): myprogram. Once I run it I get the
> message: insufficient observations to compute bootstrap standard
> errors no results will be saved. I have tried to solve the problem
> using also but it doesn't work neither. I have read STATA Manual and
> online examples but I haven't been able to understand my problem. 

Two comments:
1) You almost never want to bootstrap _se. If you do, then you want to
know the standard error of the standard error, while typically all we
want to know is the standard error of our coeficients (_b). 

As an aside, the fact that you can bootstrap the standard errors is 
nice from a pedagogical point of view: it reminds us that our measures 
of the uncertainty of our estimates are themselves also estimates and 
thus uncertain.

2) The easiest solution is probably to specify the -vce(bootstrap)- 
option in -biprobit-, like so:

*---------------- begin example ---------------
webuse school, clear
constraint 1 _b[athrho:_cons] = 0
biprobit private vote logptax loginc years, ///
         constraint(1) vce(bootstrap, reps(100))
*----------------- end example -----------------
(For more on examples I sent to the Statalist see: 
http://www.maartenbuis.nl/example_faq )

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/