|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Re: unobserved heterogeneity, clarification and confirmation
|
From |
"Rodrigo A. Alfaro" <[email protected]> |
|
To |
<[email protected]> |
|
Subject |
st: Re: unobserved heterogeneity, clarification and confirmation |
|
Date |
Wed, 27 Jun 2007 00:50:50 -0400 |
Can you fly without wings?... then you cannot improve your weak instruments
with a particular method. It is true that LIML, or LIMLF (Fuller) are robust
under weak and/or many instruments. But they are not robust under
heteroskedasticity and many instruments. Using -ivreg2- you could check a
variety of estimators that will help you in this: Donald-Newey
bias-corrected 2SLS, LIML, LIMLF (with 1 for asymptotic unbiasedness). If
you ask applied persons they will tell you to find a better set of
instruments, that is for sure "the best method".
Rodrigo
4- if instruments are weak now, can I just correct this weakness by
estimating my reg with maximum likelihood? if yes, what is the best
method?
*
* 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/
