Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: xtivreg2 vs ivregress 2sls and clustering: missing SEs in first stage

From   Francesco Amodio <>
To   statalist <>
Subject   st: xtivreg2 vs ivregress 2sls and clustering: missing SEs in first stage
Date   Thu, 17 Oct 2013 15:01:09 +0200

Dear all,

I am implementing an IV strategy where I instrument one endogenous
regressor with two exogenous regressors.

My dataset is quite big and I am interested in exploiting variation
within very fine categories. Indeed, I have around 25,000 observations
and around 2,000 categories. I thus need to have one dummy for each of
the 2,000.

In order to speed up the process, I use the 'xtivreg2' command. Once I
define my category variable using the xtset command, I can use the
'fe' option and get my estimate fast enough.

On top of this, I cluster my standard errors.

Now, I was interested in checking that my results using the 'xtivreg2'
commands where the same as the ones I obtain using the 'xi: ivregress
2sls' command.
First and second stage point estimates of the variables of interest
are exactly the same. However, when using 'ivregress 2sls' and the
clustering option, STATA is unable to produce standard errors for my
exogenous regressors in the first stage regression. In fact, it does
not display any standard errors at all. That does not happen when I do
not cluster SEs. Second stage estimates and standard errors are equal
to the ones I get using 'xtivreg2'.

Any hint on what is going on with missing SEs in the first stage regression?

Thank you very much.

Best Regards,


Francesco Amodio

GPEFM - Departament d'Economia i Empresa
Universitat Pompeu Fabra
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index