## $Id: README,v 1.1 1999/10/13 00:50:09 saikat Exp $

	     Transcripts of lme analyses of examples from
		    "SAS System for Mixed Models"

			    Douglas Bates
	     Dept. of Statistics, U. of Wisconsin-Madison

			   Jose C. Pinheiro
		    Bell Labs, Lucent Technologies

This directory contains transcripts of analyses of some of the
examples from the book "SAS System for Mixed Models" by Littell,
Milliken, Stroup, and Wolfinger (SAS Institute, 1996).  These
alternate analyses were done in S-PLUS 3.4 using version 3.0 of the
nlme library.  They primarily use the lme function for fitting linear
mixed effects models.

For each example there is a file with a .St extension which is an
S-transcript file.  Another file with a .q extension contains the
commands and comments only.  If you have the data files available as S 
objects in the directory ../data/.Data, you should be able to
reproduce the HR.St transcript, for example, by defining a .First
function so that the NLME/.Data and the SAS_Mixed/.Data directories
get attached and so that the NLME/NLME.so shared object gets
dyn.loaded.   We also use a setting for the options of
 options( width = 65, length = 10000, digits = 5, prompt = "S> " )

Then
 Splus BATCH HR.q HR.out
should produce a file HR.out similar to the HR.St file.  The graphics
will be in a file with a name like ps.out.0001.ps

An alternative, if you use ESS (Emacs Speaks Statistics) within the
emacs editor is to start an Splus process, attach the appropriate
directories, dyn.load the shared object and set the options then find
the .q file and used C-c C-n to step through the file sending the
commands to the S-PLUS process.  Of course this can also be done in
ESS by finding the transcript file itself then using C-m in the
transcript file.  ESS is available from 
		  http://franz.stat.wisc.edu/pub/ESS




