Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: bootstrap rare events


From   "Clive Nicholas" <Clive.Nicholas@newcastle.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: bootstrap rare events
Date   Thu, 2 Nov 2006 19:00:03 -0000 (GMT)

Silvia Forni wrote:

> I want to estimate bootstrap CI of parameters estimated by a logistic
> regression.
> The event i'm studing is quite rare (2%)and so in many bootstrap
> replications (96/100) paramenters could not be estimated.
>
> I increased the number of replication (5000)to obtain 300 completed
> replications, but i'm not sure it's the right method.
> Does somebody know if there are other methods?

I can't advise on whether or not it's sensible to -bootstrap- after
-logit-, as I've never tried. However, there is an alternative to
rare-events logistic modelling in -logit-: you can also try Gary King and
Langche Zeng's -relogit-, available via

http://gking.harvard.edu/stats.shtml#relogit

and then click on the -relogit- Stata version. There are two downsides to
using this:

(1) it's not the latest version of this package outright (there's a more
    up-to-date version available as part of his Zelig package, which runs
    on R);

(2) its model summary statistics are nowhere near as comprehensive as in
    -logit-.

However, to stress, I don't know if -bootstrap- would run properly with
-relogit-, but try and see.

CLIVE NICHOLAS        |t: 0(044)7903 397793
Politics              |e: clive.nicholas@ncl.ac.uk
Newcastle University  |http://www.ncl.ac.uk/geps

Whereever you go and whatever you do, just remember this. No matter how
many like you, admire you, love you or adore you, the number of people
turning up to your funeral will be largely determined by local weather
conditions.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index