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]
Re: st: plain and robust st. errors below each other with estout, etc.
From
Stas Kolenikov <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: plain and robust st. errors below each other with estout, etc.
Date
Fri, 7 Mar 2014 09:36:10 -0600
It is difficult within Stata's estimation results paradigm to have
different types of standard errors. One exception are survey settings
where there's a traditional measure of the design effect that calls
for computing both the "right" (~robust) standard errors and the
"wrong" (plain) standard errors. You can declare your data to be
svyset _n
to produce the robust standard errors (sans some of the degrees of
freedom/small sample correction multipliers), and then you can pick
the robust standard errors from e(V), and the naive standard errors
from e(V_SRS). However, I don't really know if there is a smart way to
have -estout- or -outreg- pick up these different matrices and align
them with the default _b[] and _se[] -- I can't rule out the
possibility that this is doable, but I don't really know whether they
are able to place that coefficient exactly into that location.
If I had a need like this, I would simply use -putexcel- in Stata 13
and won't bother with the user-written commands, however useful they
are for the standard tasks.
-- Stas Kolenikov, PhD, PStat (ASA, SSC)
-- Principal Survey Scientist, Abt SRBI
-- Opinions stated in this email are mine only, and do not reflect the
position of my employer
-- http://stas.kolenikov.name
On Fri, Mar 7, 2014 at 9:20 AM, Maria Boutchkova <[email protected]> wrote:
> Dear Statalisters,
>
> Usually I report my regression results in a publishable quality table
> by using estout, esttab, ourreg, outreg2, etc.
> I ran into a problem with reporting plain standard errors in
> parentheses below coefficients and then robust ones in brackets below
> the plain ones.
> I imagine it can be done by computing one set of errors yourself and
> then using estadd to store the vector, but this is prohibitively
> complicated to ask students to do.
>
> Thank you!
>
> Maria Boutchkova
> Lecturer in Finance
> University of Edinburgh
> *
> * 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/
*
* 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/