Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: dots and non-dots


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: dots and non-dots
Date   Mon, 6 Aug 2012 07:25:06 -0500

When running the replicate variance estimation methods, such as
bootstrap or jackknife, we are being entertained by the running dots
for each replication. If everything goes well, we see just the dots.
If things don't go that well, we see some red "x"s or "e"s or "s".
What are the meanings of these special codes? (Hint: I know what to
-viewsource- here, but I would hate to second guess how this cute
little fully undocumented command is being used. By "fully
undocumented" I mean that it is not even mentioned in -help
undocumented-.)

Some of the reasons when the user needs to know something went awry
may be: no observations (somehow we bootstrapped a bad sample that
does not have anything satisfying the "if" condition of the main
command); some of the estimated coefficients don't exist (-rmcoll-
dropped a different variable in the bootstrapped sample than in the
first call); an iterative procedure did not converge or produced some
other meaningless result (a perfect prediction in -logit-). There may
be other circumstances. Are these symbols explained anywhere?

If it is not documented, may be it is worth writing a blog entry about
these dots, just to help something to lean on.

-- 
---- Stas Kolenikov
-- http://stas.kolenikov.name
---- Senior Survey Statistician, Abt SRBI
-- Opinions stated in this email are mine only, and do not reflect the
position of my employer
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index