Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Nelson, Carl" <chnelson@illinois.edu> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st: Variable in bootstrap program not found |
Date | Thu, 3 Mar 2011 07:01:18 -0600 |
I am writing a bootstrap to obtain standard errors after two-step estimation with a generated regressor. When I try to execute the program I receive the following error. use full_inv, replace . capture program drop twostepseboot . program define twostepseboot, rclass 1. version 10 2. . quietly: reg cdmunicipio proprietrioa posseiroa arrendatrioa assentadoapelo > pnra larea cooperativa associacao /// > malet1 nmembrosdafamilia nempregadospermanentes literate secondary primary t > echnical superior 3. quietly: predict mun_res 4. quietly: poisson rendaactivities proprietrioa posseiroa arrendatrioa ass > entadoapelopnra larea cooperativa /// > associacao malet1 nmembrosdafamilia nempregadospermanentes literate secondar > y primary technical superior mun_res 5. quietly: local coeff proprietrioa posseiroa arrendatrioa assentadoapelo > pnra larea cooperativa /// > associacao malet1 nmembrosdafamilia nempregadospermanentes literate secondar > y primary technical superior mun_res /// > _cons 6. foreach c of local coeff { 7. scalar b`c' = _b[`c'] 8. return scalar bs`c' = b`c' 9. } 10. drop mun_res 11. end . . twostepseboot [technical] not found r(111); When I run the code through line 8 as a do file and add scalar list I see all the b`c' and bs`c' variables have been created, thus I am puzzled by the error message. And insight would be appreciated. Carl Nelson University of Illinois * * 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/