I've corresponded a bit with Weisberg at Minnesota, and have
received some updates to r-code that permit fitting glims.
I've also tracked down the documentation for lisp-stat -- I've
attached the information. My hope is that this document might be
a useful course handout (I think that it's more or less the second
chapter of Tierney's book.)
Anyway, if you're interested, here's the info. Note that the info is
old, and doesn't describe the most recent release of lisp-stat, but it
does explain how to extract the documentation.
Take care,
John
Release 2 of XLISP-STAT Version 2.1. Released February 1991.
Lisp-Stat is an extensible environment for statistical computing and
dynamic graphics based on the Lisp language. XLISP-STAT is a version
of Lisp-Stat based on a dialect of Lisp called XLISP.
Originally developed for the Apple Macintosh, XLISP-STAT is now also
available for UNIX workstations using the X11 window system and for
use on Sun workstations running suntools. A version for the Commodore
Amiga is available as well. To facilitate statistical computations
standard Lisp functions for addition, logarithms, etc., have been
modified to operate on lists and arrays of numbers, and a number of
basic statistical functions have been added. Many of these functions
have been written in Lisp, and additional functions can be added
easily by a user. Several basic forms of plots, including histograms,
scatterplots, rotatable plots and scatterplot matrices are provided.
These plots support various forms of interactive highlighting
operations and can be linked so points highlighted in one plot will be
highlighted in all linked plots. Interactions with the plots are
controlled by the mouse, menus and dialog boxes. An object-oriented
programming system is used to allow menus, dialogs, and the response
to mouse actions to be customized.
If you have access to anonymous ftp over the internet you can obtain
the most recent version of the XLISP-STAT source code for the
UNIX and Macintosh versions from umnstat.stat.umn.edu (128.101.51.1).
Executable versions for the Macintosh are available by anonymous ftp
as well. All files are located in the directory pub/xlispstat.
IF YOU CAN FTP THE FILES, PLEASE DO THAT. Ftp'ing the files will be
much easier for you and will not place such a big burden on statlib
itself.
If you do not have access to anonymous ftp you can obtain the source
code and documentation for XLISP-STAT from statlib. There are two
files,
xlispstat.src - uuencoded compressed tar file of the source code
xlispstat.doc - uuencoded compressed tar file of LaTeX source
These files are approximately 1200K and 200K in size, respectively. You
will need the utilities uudecode, uncompress and tar to unpack them.
To obtain the xlispstat files send the messages
send xlispstat.src from xlispstat
and
send xlispstat.doc from xlispstat
to
statlib@lib.stat.cmu.edu. You will receive each file split in a large
number of small parts, along with instructions on assembling the
parts.
After you have assembled the split files, you will need to uudecode,
uncompress, and un-tar them. If you have requested the xlispstat.doc
file and if the information has been reassembled into a file called
(say), wilma, then on many systems the following commands will work:
uudecode wilma
will create a file called xlispstat.doc.tar.Z
At that stage the command
zcat xlispstat.doc.tar.Z | tar -xvf -
will uncompress and un-tar the information.
The technical report in xlispstat.doc contains a tutorial introduction
to the system. More complete documentation will be published as a book
that should be available from Wiley in October 1990. The title of the
book is "Lisp-Stat: An object-oriented environment for statistical
computing and dynamic graphics."
For further information contact
Luke Tierney
School of Statistics
University of Minnesota
Minneapolis, Mn. 55455
luke@umnstat.stat.umn.edu