[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
"Powers, Nicholas" <Nicholas.Powers@brattle.com>
"'firstname.lastname@example.org' (email@example.com)" <firstname.lastname@example.org>
Sat, 29 Oct 2011 16:30:20 -0400
I am wondering if, in the course of Stata's -xtregar- command, or some user-written routine, it is possible to obtain the Baltagi and Wu (1999) transformation of the data (the Baltagi and Li 1991 version would also work, as my dataset is a strongly balanced panel of 48 states over 32 years). I understand that -xtregar- performs the transformation in the course of executing, and then runs OLS on the transformed data, but without the option to specify other variance options (i.e. robust, cluster, bootstrap).
I'd like to recover the transformed data so that I can run OLS and obtain more conservative estimates of the variance; while I understand that -xtregar- will remove any persistent autoregressive component in the error term as well as the unit-specific component of the residuals, it seems to me that there could still be heteroskedasticity across units or transitory correlation over time within units, so I'd like to re-run OLS with a more flexible variance structure.
Looking online and through Statalist has not helped as of yet. Any help would be appreciated. For reference, I have a balanced panel of 48 states over 32 years.
Nicholas E. Powers
The Brattle Group
1850 M Street NW, Suite 1200 | Washington, DC 20036 | United States
Main: +1.202.955.5050 | Fax: +1.202.955.5059
Cambridge | San Francisco | Washington | Brussels | London | Madrid
IMPORTANT NOTICE FROM THE BRATTLE GROUP: This message, and any attachments, are
intended only for the use of the individual or entity to which it is addressed and
may contain information that is privileged, confidential, and exempt from disclosure
under applicable law. Any unauthorized dissemination, distribution or copying of this
communication is strictly prohibited. If you have received this communication in error
please notify me immediately by return email and immediately delete the original and all
copies of the message and any attachments to it. Note also that nothing in this message is
intended to constitute an electronic signature or otherwise to satisfy the requirements for
a contract unless an express statement to the contrary is included in the message.
Please ensure you have adequate virus protection before you open or use attachments.
The Brattle Group does not accept any liability for viruses.
* For searches and help try: