Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: bootstrap and conformability error


From   "sandrine rospabe" <srospstata@hotmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: bootstrap and conformability error
Date   Mon, 28 Jul 2003 00:27:44 +0000

Dear statalist,

I am trying to estimate the significance of a few statistics using a bootstrap analysis.

I have created a program in which I estimate for 2 different sub-samples (respectively 6610 and 1114 observations) a regression with a treatment effect (treatreg command) using the MLE and then compute the difference in expected earnings between those who receive the treatement and those who don't. Having done this for the 2 sub-samples I compute the difference between the treatment earnings effect for subsample1 minus the treatment earnings effect for subsample2.
Then I use the command bstrap to test for the significance of these 3 differentials.

My problem is the following: When I set 50 iterations (reps50), the program is working. But when I set 200 iterations (which is the number of iterations I need), an error message says "conformability error".

I guess the problem comes from the drawing of a particular sample which prevents a certain matrix computation. But I just don't know how to solve the problem.

Would someone have an idea of a solution?

Many thanks

Sandrine Rospabe


Dr. Sandrine Rospabé
Université de Rennes I - Faculté des Sciences Economiques
CREREG
7 Place Hoche
35065 Rennes cédex
02-23-23-33-24
06-24-24-07-63

_________________________________________________________________
Découvrez les nouvelles émoticônes animées de http://g.msn.fr/FR1001/866 MSN Messenger nouvelle formule

*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* 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   |   What's new   |   Site index