re: st: reg3 option: -robust-

[Christopher F Baum <baum@bc.edu>]

<>
John said

You could use the -boostrap- option with reg3. Bootstrapped SEs are not 
that different from -robust- ones. See:

Ando, M., & Hodoshima, J. (2007). A note on bootstrapped White's test 
for heteroskedasticity in regression models. Economics Letters, 97(1), 
46-51.

If memory serves me right, they talk about the general regression case 
(and not simultaneous equations), but this should make no difference to 
the reg3 case.




However to my knowledge there is no reason, in terms of econometric theory, why -sureg- and -reg3- cannot provide vce(robust) or (vce(cluster cvar) options. If there is, I'd like to hear about it. If there isn't, I'd like to see these routines updated. The fact that the more-recently-developed -nlsur- provides both options (but could be used to estimate a linear SUR) suggests that there is no obstacle for -sureg-, and -reg3- is very similar to -sureg- in its logic. I have complained about this many times in 'wishes and grumbles' sessions and on this list, to no avail.

KIt

Kit Baum   |   Boston College Economics and DIW Berlin   |   http://ideas.repec.org/e/pba1.html
An Introduction to Stata Programming   |   http://www.stata-press.com/books/isp.html
An Introduction to Modern Econometrics Using Stata   |   http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/