Statalist


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

Re: st: boostrap CI for intraclass correlation coefficient


From   "Stas Kolenikov" <[email protected]>
To   [email protected]
Subject   Re: st: boostrap CI for intraclass correlation coefficient
Date   Thu, 17 Apr 2008 10:33:57 -0500

Oh come on, why would you want to bootstrap this? If you have the
estimates of variances along with their standard errors, then you can
compute the estimated variance of estimated ICC through the delta
method. It is easily implementable in Stata through -nlcom-.

In fact it will be more difficult to set up the bootstrap properly, as
the bootstrap for mixed models is not an easy task... and in some
situations like two-way models, no bootstrap can give a consistent
estimate of the true variances (see
http://www.citeulike.org/user/ctacmo/article/2682537,
http://www.citeulike.org/user/ctacmo/article/2682551)

On 4/17/08, Eddy Fan <[email protected]> wrote:
> Hello,
>
>  I have estimated the ICC in strength measurements between raters on a
>  serious of patients by using xtmixed (for the variance components) in
>  the following way:
>
>  xtmixed d_strength || rater_id:
>
>  Then I get estimates of the variances and their standard errors.  I
>  then calculated ICC as the proportion of total variance that occurs
>  between raters.  How do I bootstrap to get 95% confidence intervals
>  for the calculated ICC?
>
>  Thanks.
>
>  Eddy
>
>
>  *
>  *   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/
>


-- 
Stas Kolenikov, also found at http://stas.kolenikov.name

Small print: Please do not reply to my Gmail address as I don't check
it regularly.
*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index