[Joao Santos Silva]

Dear All,

With the usual thanks to Kit Baum, an updated version of xtqreg is now available in SSC. This version allows factor variables and solves a small issue with bootstrapping the command.

Please do let me know if you find any problems with the updated command.

Best wishes,

Joao

[Jon Burzka]

Dear Joao Santos Silva

I conducted a test regarding serial correlation and heteroscedasticity which confirms that both are present. now I wonder if this setting is covering both issues with clustered-robust-SE:

bootstrap, reps(50) seed(20) cluster(year) idcluster(id4): xtqreg $ylist $xlist, i(year) quantile(.25)


Best
Jon

[Joao Santos Silva]

Clustering by year is very unusual; are you sure you want to do that? Anyway, I do not think the way you are doing it accounts for serial-correlation.

Joao

[Jon Burzka]

Thank you.

You are correct, it doesn't account for serial correlation.
Though when I change the cluster to id, the results seem to be flawed..

Best
Jon

[Joao Santos Silva]

in xtqreg, you are not clustering by the right variable.

Joao

[Nicola Rossi]

Dear Professor Santos Silva, I am trying to create tables launching eststore and then outreg2, but it only gives me back in xls the estimates regarding the last quantile. Is there any way to obtain all the estimates?

Thank you in advance.

All the best

[Joao Santos Silva]

Dear Nicola Rossi,

I am afraid that at the moment there is no way to do it; you have to estimate each quantile separately. That is something I can put in my to-do list.

Best wishes,

Joao