Does anyone have an idea how to implement condivreg
and compute moreira size-adjusted tests (condtest,
newtest) including an individual-dummy/fixed effect
and be sure it yields computes tests that adequately
adjust for the fewer degrees of freedom in an FE/FD
regression?
I try it by first converting the data using xtdata,
fe, and then running condivreg on this data, and then
'condtest' (or actually "newtest", a newer version).
It runs fine but I am not sure if the resulting tests
consider the degrees of freedom lost from estimating
(in my case, over 4000) fixed effects. Any thoughts
on how to adapt this or make the correction necessary
to account for the loss of degrees of freedom?
I confess I am not well-versed on precisely how theses
size-adjusted tests are computed, but I now they are
supposed to be more robust to the possibility of a
weak instrument.
Thanks for your help,
David Reinstein
UC Berkeley
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
*
* 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/
