To my knowledge, the defining works in this area are the following:
Mixed Effects Models in S and S-Plus
by Jose C. Pinheiro, Douglas M. Bates (Hardcover)
Amazon Avg. Customer Rating: 5.0 out of 5 stars
Nonlinear Regression Analysis and Its Applications
by Douglas M. Bates (Author), Donald G. Watts (Author) (Hardcover -
August 1988)
Amazon Avg. Customer Rating: 5.0 out of 5 stars
Doug Bates' PhD dissertation over 20 years ago established that the
primary problem with nonlinear estimation was "parameter effects
curvature". In almost all cases, it is much a much bigger problem than
"intrinsic curvature". In "Nonlinear Regression", Bates and Watts show
how appropriate reparametization can make nonlinear estimate and the
accompanying asymptotic normal theory much more effective and useful.
"Nonlinear Regression" is a seminal contribution to statistical science,
and "Mixed Effects Models" is a detailed manual essential for effective
use of nlme in S-Plus.
The function "optim" in the MASS library is useful with nonnormally
distributed responses.
Hope this helps.
Spencer Graves
Unspecifed wrote:
Dear All,
I am a novice of S-Plus.
Would someone recommend good books on non-linear models using S-Plus.
I want to write programmes for the estimation of non-linear simultaneous
equations by S-Plus.
Thank you in advance.
Harry Wu
--------------------------------------------------------------------
This message was distributed by s-news@lists.biostat.wustl.edu. To
unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
the BODY of the message: unsubscribe s-news
|