Skip to content
Back to formatted view

Raw Message

Message-ID: <alpine.LMD.2.00.1504081336290.6246@orpheus.qimr.edu.au>
Date: 2015-04-08T03:40:58Z
From: David Duffy
Subject: bootstrapping a mixed-model (lme)
In-Reply-To: <CAJBOW-uU4J_hhYxYcqUkVtJgKQJBUxrFZaE=xr-mPZdq2EX9jA@mail.gmail.com>

On Wed, 8 Apr 2015, Roslyn Dakin wrote:

> This question is in response to Ken's point about bootstrapping mixed
> models (that you have to resample whole clusters/subjects for the
> nonparametric bootstrap).

I've played a bit with the delete-d jackknife (the book by Shao and Tu is 
good). You just delete a randomly selected subset of observations to 
generate each pseudosample - that way the underlying correlation 
structure/clustering is retained.

| David Duffy (MBBS PhD)
| email: David.Duffy at qimrberghofer.edu.au  ph: INT+61+7+3362-0217 fax: -0101
| Genetic Epidemiology, QIMR Berghofer Institute of Medical Research
| 300 Herston Rd, Brisbane, Queensland 4006, Australia  GPG 4D0B994A