It's just a caveat; neither the variance inflation of Kackar and Harville (84) nor the t-statistics degree of freedom juggling that SAS uses are being applied, so the variance is understated relative to those. My understanding is that it makes little difference in many cases in practice, and I have no idea if the coverage properties perform as advertised. Ryan King Dept Health Studies University of Chicago On Mon, Nov 22, 2010 at 5:00 AM,
<r-sig-mixed-models-request at r-project.org> wrote:
Date: Sun, 21 Nov 2010 18:11:54 +0100 From: Petar Milin <pmilin at ff.uns.ac.rs> To: Dimitris Rizopoulos <d.rizopoulos at erasmusmc.nl> Cc: R-SIG-LMER <r-sig-mixed-models at r-project.org> Subject: Re: [R-sig-ME] Are prediction interval values ungettable? Message-ID: <4CE952DA.1020208 at ff.uns.ac.rs> Content-Type: text/plain; charset=ISO-8859-1; format=flowed This is right, but I have compared values I get with the below procedure explained by Andrew and the end points of whiskers on the caterpillar plot (I tried to approximate them the best I could), and they do match for five different data sets and models. Best, PM