### Tuesday, February 28, 2006

## st: RE: RE: RE: fe & robust

thanks mark so much for your useful comments. one more question . if the command of fe by using cluster is

areg ldiv1 lgov1 lfc size lsale llev lbeta2 lroe if country==1, absor(company) cluster (company)

what is the command of random effects by using cluster - for the same variables- should be

thanks duha

>>> M.E.Schaffer@hw.ac.uk 28/02/2006 12:02:44 >>> Duha,

> -----Original Message----- > From: owner-statalist@hsphsun2.harvard.edu > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of > Duha Al-Kuwari > Sent: 28 February 2006 11:41 > To: statalist@hsphsun2.harvard.edu > Subject: st: RE: RE: fe & robust > > hi mark, thanks a lot for your reply... > > i attached the result , would you please have a look and > tell me your comment > > best regards > duha

Two things. First, it's not good Statalist practice to use attachments. Put the output in the text of your email next time. Second, my guess about cluster was correct - areg is using -cluster- and xtreg is not. That's why the coefficient estimates are the same but the standard errors are different.

Cheers, Mark

> > > > > hi again....there was a small mistake in my previous > question....here > > is my corrected one: > > > > i used two commands to test fixed effects model with robust. > > the first one is > > : > > > > xtreg ldiv1 lgov1 lfc size lsale llev lbeta2 lroe if > > country==1, fe > > robust > > > > the second one is: > > > > areg ldiv1 lgov1 lfc size lsale llev lbeta2 lroe if country==1, > > absor(company) cluster (company) > > > > but results of the two commands were different although > > they are testing > > fixed effects model...does any one know the reason > > > > thanks > > duha > > It's still not obvious they are estimating the same model. > The areg estimation uses company as the observational unit > and clusters on company; the xtreg estimation does not use > -cluster-, and you don't say what the observational unit is. > > It would help if you included the first few lines of the > estimation results for the two commands. > > --Mark > > > > > > > >>> M.E.Schaffer@hw.ac.uk 28/02/2006 10:41:51 >>> > > Duha, > > > > > -----Original Message----- > > > From: owner-statalist@hsphsun2.harvard.edu > > > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Duha > > > Al-Kuwari > > > Sent: 28 February 2006 10:28 > > > To: statalist@hsphsun2.harvard.edu > > > Subject: st: fe & robust > > > > > > hi all > > > > > > i used two command to test fixed effects model with robust. > > > the first one is > > > : > > > > > > xtreg ldiv1 lgov1 lfc size lsale llev lbeta2 lroe if > > > country==1, re > > > robust > > > > > > the second one is: > > > > > > areg ldiv1 lgov1 lfc size lsale llev lbeta2 lroe if country==1, > > > absor(company) cluster (company) > > > > > > but results of the two commands were different...does any > > > one know the > > > reason > > > > > > thanks > > > duha > > > > The -re- option for xtreg means you estimated a random > effects model > > (unless this was a typo in your email). > > > > --Mark > > > > > > > > * > > > * 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/ > > > > > > > > > > * > > * 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/ > > * > > * 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/ > > > > > > * > * 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/ > >

* * 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/ * * 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/

Tag: statalist