Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: RE: error code from bssize refine


From   "Brian P. Poi" <bpoi@stata.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: error code from bssize refine
Date   Thu, 24 Jan 2013 08:39:42 -0600

On 01/24/2013 05:35 AM, Murat Genc wrote:
Dear Statalist,

I am using -bssize- from SJ 11-4 in Stata 12.1. All is well with the following three lines of commands:

	bssize initial, tau(0.01) pdb(5)
	set seed 1
	bootstrap, reps(1326) saving(one): sureg (`sequat1') (`sequat2')(`sequat3') 	(`sequat4'), c(1/`k')

I can see the coefficients and their bootstrapped standard errors for all the equations I estimate.

However, when I next type the command

	bssize refine using one

I get the error message

	no valid variables found
	r(498);

I fail to understand why I get an error.  I am following the example in the SJ paper I referred to above. The set of commands I use is identical to the ones in the paper (and also to the ones in the help file).

I searched the usual sources but I couldn't find the answer.  I'll be grateful if someone can help.

Murat


Murat,

The short answer is that -bssize- just is not up to handling multple-equation estimation commands as it is now written.  When you use -bootstrap- with a multiple-equation command, the names of the variables in the results dataset begin with an equation name, and -bssize refine- is not set up to deal with variables named like that.  I will see if I can get this fixed at some point.

   -- Brian Poi
   -- bpoi@stata.com

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index