st: Re: ivreg2/xtivreg2 and two-way clustering syntax

[Wafa Hakim Orman <wafa1024@gmail.com>]

Hi everyone,

I'm using Stata 10.2 and trying to run a regression with two-way
clustered standard errors. I've run update ado, adoupdate, & all the
other update commands. Still, when I try two-way clustering,
Driscoll-Kraay standard errors, or any other such new option it throws
an error. Here is my output:


. ivreg2 charges quarter year unemployment , cluster(statefips qtryr)
cluster():  too many variables specified
r(103);


. ivreg2 charges quarter year unemployment , gmm2s cluster(statefips year)
cluster():  too many variables specified
r(103);

. xtivreg2 charges quarter year unemployment , fe gmm2s cluster(statefips year)
cluster():  too many variables specified
r(103);

. ivreg2 charges quarter year unemployment , bw(2) cluster(qtryr)
cannot use HAC kernel estimator with -cluster- option
r(198);

. ivreg2 charges quarter year unemployment , dkraay(1)
option dkraay() not allowed
r(198);


My dataset is a panel, with statefips as the panel ID & qtryr as the
time variable. What am I doing wrong? Am I missing something obvious
in the syntax?


I appreciate any help Statalisters can provide!

Thanks in advance,
Wafa.

--
"So be it."
                  --Kurt Vonnegut
*
*   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/