Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Austin Nichols <austinnichols@gmail.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: Regression with clustered robust standard errors |
Date | Tue, 11 Feb 2014 09:19:23 -0500 |
Alberto Valzolgher - 1577733 <alberto.valzolgher@studbocconi.it>: If you want to cluster by year and industry (2 dimensions), use -ivreg2-: ssc inst ivreg2 help ivreg2 and read http://www.stata.com/meeting/uk10/UKSUG10.Baum.pdf On Tue, Feb 11, 2014 at 7:22 AM, Nick Cox <njcoxstata@gmail.com> wrote: > You are correct and your supervisors are wrong. > > The syntax is clear: Stata expects a variable name. > > Nick > njcoxstata@gmail.com > > > On 11 February 2014 10:45, Alberto Valzolgher - 1577733 > <alberto.valzolgher@studbocconi.it> wrote: >> Dear statalist, >> >> I am conducting a research on stock market first day return, and I'd like >> to regress with clustered robust standard errors, clustering by year and >> industry. >> >> For this purpose I am using the option vce(cluster YEAR). My question is, >> does the system consider just one cluster (YEAR) or it considers the n >> years in the YEAR variable as clusters (in my output is written s.e. adj. >> for 11 clusters in YAER)? >> I think the second interpretation is correct but my supervisors told me i >> should specificate in the cluster option all the clusters e.g. cluster >> (2001 2002 2003...). * * 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/