The nlme software has been completely redesigned to emphasize a
modular, object-oriented design that facilitates users' incorporating
new methods for existing generic functions (e.g. new classes of
correlation structures and variance functions). Some of the new
features of nlme 3.0 are:
- Multilevel models for data with multiple nested grouping levels.
These use new computational methods described in (PostScript)
http://franz.stat.wisc.edu/pub/NLME/CompMulti.ps
or (Adobe Portable Document or "Acrobat" Format)
http://franz.stat.wisc.edu/pub/NLME/CompMulti.pdf
- a new class, called groupedData, for representing data grouped
according to one or more nested factors. Methods for plotting,
summarizing, and fitting groupedData objects are also included.
- extensive use of Trellis plots for exploring grouped data and
checking models fitted to such data. The plot methods for fitted
objects include a formula argument which gives them unlimited
flexibility for generating diagnostic plots.
- new classes of correlation structures including ARMA(p,q) models,
spatial correlation structures (Gaussian, exponential, etc.) with and
without nugget effects, general correlation with no particular
structure, and a Huyn-Feldt structure.
- new and redesigned classes of variance functions using arbitrary
covariates and grouping of parameters (the previous version used the
fitted values as the covariate for the built-in variance
functions). Because the correlation structure and the variance
function are independently specified, heterogeneous AR1, ARMA, etc.
models are handled naturally.
- a gls function for fitting linear models with correlation structures
and/or variance functions. You can think of it as lme without random
effects.
- an extensive set of examples. A separate directory contains
groupedData objects for the datasets used in the book "SAS System for
Mixed Models", by Littel, Milliken, Stroup, and Wolfinger. Sample lme
analyses that parallel the PROC MIXED analyses from that book are also
given. We hope this will enable people who are familiar with PROC
MIXED to learn lme more quickly.
Those interested in becoming beta testers can obtain the code from
http://cm.bell-labs.com/stat/project/nlme/Beta
ftp://cm.bell-labs.com/cm/ms/departments/sia/project/nlme/Beta
or
http://franz.stat.wisc.edu/pub/NLME/Beta
ftp://franz.stat.wisc.edu/pub/NLME/Beta/
The Unix version is available as a gzip'd tar file
(nlme3_0b1.tar.gz). The Windows version is available as a zip'd file
(nlme30b1.zip). The datasets and sample lme analyses for the examples
in the "SAS System for Mixed Models" book are included as a separate
zip'd file (mixed.zip) in the Windows version. If you use ftp to
transfer one of these files, remember to set binary mode for the
transfer.
If you decide to test the code, we strongly recommend you subscribe to
nlme-announce@stat.wisc.edu and also to nlme-help@stat.wisc.edu. Send
a message with the word "subscribe" in the body to
nlme-announce-request@stat.wisc.edu and to
nlme-help-request@stat.wisc.edu. The first list should be very low
traffic as it will just provide announcements of new beta versions
(can you say "bug fixes"?). The second is the recommended list for
beta-testers to ask questions about why things don't work the way they
expect them to.
-- Jose' Pinheiro jcp@research.bell-labs.com Bell Laboratories (908) 582-2390 Lucent Technologies http://cm.bell-labs.com/stat/jcpand
Douglas Bates bates@stat.wisc.edu Statistics Department 608/262-2598 University of Wisconsin - Madison http://www.stat.wisc.edu/~bates/
----------------------------------------------------------------------- This message was distributed by s-news@wubios.wustl.edu. To unsubscribe send e-mail to s-news-request@wubios.wustl.edu with the BODY of the message: unsubscribe s-news