Bookmark and Share

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


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

st: RE: "cluster(): too many variables" in ivreg2 despite latest version?


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: "cluster(): too many variables" in ivreg2 despite latest version?
Date   Thu, 17 Oct 2013 12:33:55 +0000

JZ,

Re: (1), exactly what does -which ivreg2, all- report?  When you say you downloaded the "2013 versions", from where you did download them and how?  And have you also downloaded the Mata library file livreg2.mlib?

--Mark

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of Jen Zhen
> Sent: 17 October 2013 13:01
> To: statalist@hsphsun2.harvard.edu
> Subject: st: "cluster(): too many variables" in ivreg2 despite latest version?
> 
> Dear listers,
> 
> (1) When specifying two-dimensional clustering in -ivreg2-,
> ivreg2 depvar controlvar (endogreg = instrument), cluster(dimension1
> dimension2), I always get the error "cluster(): Too many variables specified".
> 
> Many earlier posts on the topic on Statalist suggest that this problem should be
> solved by ensuring to have at least the 2010 version of -ivreg2-. While my Stata
> cannot be updated automatically, I have downloaded 2013 versions of both
> ivreg2 and of ranktest, and I have also checked -which ivreg2, all- that I
> currently have only that 1 version of the ado file in my PLUS ado file path.
> Nonetheless the problem persists. Would anyone know what else I could try?
> Does ivreg2 maybe require any other ado files that might also need manual
> updating?
> 
> (2) Nichols 2007 recommends to have at least 50 different clusters and Angrist
> & Pischke similarly suggests to have at least 42. So I'm taking this into account
> when choosing on which dimensions to cluster my standard errors. What I'm
> wondering though is: Should I also make sure to have no or few clusters with
> fewer than ... observations? In particular, with the way I am currently
> clustering, I have a few clusters with just 2 observations, although 95% of my
> clusters have more than 17. With 2 way clustering, any such requirement would
> presumably need to be independently satisfied for each of my 2 dimensions?
> 
> Thank you so much and kind regards, JZ
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/


----- 
Sunday Times Scottish University of the Year 2011-2013
Top in the UK for student experience
Fourth university in the UK and top in Scotland (National Student Survey 2012)

We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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