[R-sig-ME] Problems withestimationg model in lmer

[Nicholas Lewin-Koh]

Hi Doug,
Actually the effect is subtle. If you look at slice by day 
and look at the dose response curves from the model there is 
a small effect at small doses in the slope, as the dose increases there is 
actual suppression.

By the way, even though I tried to strip off anything that might be
informative about which compound this is, please understand that this is
company data. I would ask that it not be distributed beyond your class. I do
not have the authority to ok its release to a broader audience.

Thanks for your understanding
Nicholas

-----Original Message-----
From: dmbates at gmail.com [mailto:dmbates at gmail.com] On Behalf Of Douglas
Bates
Sent: Thursday, February 14, 2008 10:18 AM
To: Nicholas Lewin-Koh
Cc: r-sig-mixed-models at r-project.org; Bert Gunter;
lattice-seminar at cs.wisc.edu
Subject: Re: [R-sig-ME] Problems withestimationg model in lmer

Thanks for sending the data with your report,  Nicholas.  I found it
very interesting, especially as a group of us are conducting a reading
course using Deepayan Sarkar's forthcoming book "Lattice: Multivariate
Data Visualization with R" (Springer, 2008).  (The book is due out
next month - Deepayan was kind enough to let us use a preprint copy
for the course.)  I enclose a plot of these data and the R code that
generates the plot.  It is interesting because it shows a thresholding
kind of behavior.  There is not much effect for dose until about 1.25
mg/kg.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: doseresp.pdf
Type: application/pdf
Size: 26176 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-mixed-models/attachments/20080214/ba4e1885/attachment.pdf>