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: Re: jackknifing and loop question [was: <meaningless title>]


From   Nick Cox <njcoxstata@gmail.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   st: Re: jackknifing and loop question [was: <meaningless title>]
Date   Tue, 12 Mar 2013 16:00:14 +0000

You have to think about a loop. The analogy with jackknifing here is far too tenuous to be pertinent to your programming.

But before you think about a loop, have you done the combinatorics here? For 10 provinces, I count about 1000 regressions, and naturally they are highly dependent.

Nick


On 12 Mar 2013, at 15:36, "Von Fintel, Dieter <dieter2@sun.ac.za>" <dieter2@sun.ac.za > wrote:

I have a question about delete k jackknife estimation. However, I would like to delete k groups at a time, rather than k observations.

Basically I want to run a regression of the type:

reg y  i.province othercontrols

I then want to leave out all combinations of 1 province (with a dataset saving coefficients). Then I want to leave out all combinations of 2 provinces, all combinations of 3 provinces, etc

Up to this point, all that I see would be possible is all combinations of deleting 1 observation at a time, or deleting 1 cluster at a time. I could define province as my cluster, but this will still only leave out one at a time. Other than writing a loop, is there any way that you know of how I could use the jackknife command to achieve this.
*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index