Hello,
I posted a question about the 'problem with ivreg2 with cluster option',
and Prof . Mark Schaffer gave me an advice. I just want to make clear
some details of the data I am using. Cross-sectional dimension is 50
(firms) and the time dimension is 6 years (from 1997 to 2002). There is
just one endogenous variable in the regression, and I am using two
instruments (which comply with the criteria required, i.e., they are
correlated with the endogenous variable and appear sgnificantly
different from zero in the first stage, also they are orthogonal to the
error term).
There is not much difference between the two approaches (kernel-based
robust and cluster robust) in terms of the size of coefficients and
significance. The only difference is the one I mentioned in my previous
message (the F test of the overall significance in the second stage show
that coefficients are not different from zero in the cluster robust
option).
If I got it right, the approach I should use will depend on my data.
Kernel-based approach is suitable if time dimension is long and cluster
approach is suitable if degrees of freedom (number of firms minus the
number of instruments) are not small. I will appreciate an advice,
about the most suitable approach (if any), from Prof. Mark Schaffer
again.
Thank you
Entela Shehaj
The information in this email is confidential and is intended solely for
the addressee. Access to this email by anyone else is unauthorised.
If you are not the intended recipient, any disclosure, copying,
distribution or any action taken or omitted to be taken in reliance on
it, except for the purpose of delivery to the addressee, is prohibited
and may be unlawful. Kindly notify the sender and delete the message and
any attachment from your computer.
*
* 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/
