Quantcast

st: Heteroskedasticity-robust F statistic and t statistic

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: Heteroskedasticity-robust F statistic and t statistic

Carola Herrera
Good morning,

How do I compute heteroskedasticy-robust F statistic for an OLS regression?

If I use reg with , robust (cluster)
are the t statistics obtained robust to heteroskedasticity or do I need another command?

Any help on this question will be appreciated (I have been looking at STATa help and FAQ files but I am still not clear on these distinctions)
Thanks
Carola


     

*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: AW: Heteroskedasticity-robust F statistic and t statistic

Martin Weiss-5

<>

You probably want - vce(robust)- for your varcov. With the standard errors
changing under the new estimator for the varcov, the F-test must change as
well - which you can observe in the example below.

*************
sysuse auto, clear
reg pr we tr disp
di in red "F-statistic: " e(F)
reg pr we tr disp, vce(robust)
di in red "F-statistic: " e(F)
*************



HTH
Martin

-----Ursprüngliche Nachricht-----
Von: [hidden email]
[mailto:[hidden email]] Im Auftrag von Carola Herrera
Gesendet: Donnerstag, 28. Mai 2009 22:24
An: [hidden email]
Betreff: st: Heteroskedasticity-robust F statistic and t statistic

Good morning,

How do I compute heteroskedasticy-robust F statistic for an OLS regression?

If I use reg with , robust (cluster)
are the t statistics obtained robust to heteroskedasticity or do I need
another command?

Any help on this question will be appreciated (I have been looking at STATa
help and FAQ files but I am still not clear on these distinctions)
Thanks
Carola


     

*
*   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/


*
*   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/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

st: Bloom and Killingsworth, 1985, Truncated Reg

Murat Karaoz
Hi Everyone,

I am trying to estimate truncated regression using the method offered by
Bloom and Killingsworth (1985). This method is used instead of the
conventional one, when part of the dependent variable (the ones =0) of
probit equation do not exist. This estimation method is not availabe with
stata. as far as I have checked it is available only with TSP 5.0, which I
can not find in short time.

My question is, is there anyone who developed a module in stat for the
Bloom and Killingsworth (1985) procedure for truncated regression, where
the selection variable for dependent variable is partially not observed.

Thank you very much

Murat Karaoz

references
Bloom, D., Killingsworth, M., 1985. Correcting for truncation bias caused
by a latent truncation variable. Journal of Econometrics 27, 131– 135.

the method is for instance used by:
M.G. Colomboa, M. Delmastrob, L. Grilli,2004, Entrepreneurs’ human capital
and the start-up size of new technology-based firms, International Journal
of Industrial Organization, 22  1183– 1211.


*
*   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/
Loading...