We try to keep a collection of everything connected with Xlisp-Stat
in these directories. There is always a version of the source code,
there are some binaries for selected systems. As the path to the
directory shows, there are other lisp systems besides xlisp,
and other xlisp extensions besides xlisp-stat.
The most interesting sub-directories are contrib and homegrown.
In homegrown we have Xlisp-Stat code written at UCLA. This includes
- the SIR package by Ker-Chau Li and his students,
for Slicing Inverse Regression.
- the Terrace package by James Hilden-Minton for Hierarchical Linear
Modelling with diagnostics.
- the Gifi-gardens package by James Hilden-Minton for MVA with
optimal scaling.
- many utilities by Jan de Leeuw for monotone regression, b-splines,
the basic APL array operators, multidimensional scaling, nonlinear
MVA, FORTRAN-type formatted input, log-linear modeling.cd
- stem-and-leaf-plots by (our) David Betz.
- bootstrap/jackknife algorithms by Jan de Leeuw. This includes
bootstrap-t, percentile, abc, bca confidence intervals, bootstrap
and jackknife standard error, bootstrap and jackknife bias estimation,
bootstrap prediction error estimation. The bootstrap algorithms
for regression are written as methods for regression-model-proto.
There is also a graphical bootstrap demo. And there is code
for permutation and bootstrap hypothesis testing. There are also
versions of Randy Sitter's algorithms for bootstrapping in
survey situations with clustered and stratified random samples
(without replacement).
- Didactic Modules by Jan de Leeuw. Moving the regression line,
moving points in regression scatterplots, the central limit theorem.
- UCLAData. Evolving package for Introductory Data Analysis teaching
by Jan de Leeuw.
- Help. Evolving graphical help system for Xlisp programmers and Users
by Jan de Leeuw and Jason Bond.
* Note **************************************************************
I write my Xlisp-Stat code in planes, on airports, and in hotel rooms.
If you want more useful utilities, you'll have to invite me over (or,
if you prefer that, send me somewhere else).
*********************************************************************
In contrib we have xlisp-stat code written by other. This includes
everything on statlib, and then some. We mention
- axis: Introductory statistics package by Robert Stine (Wharton).
- bpois: Baysian Poisson Regression using the Gibbs Sampler by
Hani Doss (OSU) and B. Narasimhan (PSU).
- cw: Dynamic Graphics for Regression, by Dennis Cook and Sande
Weisberg (U Minn). Called dyndiag on statlib.
- eltoy: Introductory Statistics and Bayesian Elicitation, by
Russel Almond (U Wash).
- expsurv: Exploratory Survival Analysis by K. Neely Atkinson
(U Texas).
- hasse: F tests and ANOVA tables for factorial designs by
Philip Iversen (Iowa State).
- jflisp: Three packages for Model Selection in Regression Analysis
by Julian Faraway (Univ. Michigan).
- naras: Code contributed by B. Narasimhan (see also Bpois). This
includes utilitity code for sliders and boxplots, as well as
packages to demonstrate Markov chains, and to fit kernel density
estimates. There is also a package for programmers who want to
dynamically link arbitrary C functions at runtime.
- plottools: Don't know who wrote it. It provides nice graphical
windows to selects symbols and colors.
- puranen: Some demos by Juha Puranen (Univ Helsinki). Demoes
Box-Cox and some of the common densities.
- simulation. General purpose simulation tool by Robert Stine (Wharton).
- students: Projects by students in Nancy Reid's course at the
U Toronto. Code for repeated measures with missing data, teaching
demos, correspondence analysis, Weibull MLE, the general linear
hypothesis.
- surv: Code for survival analysis by E. Neely Atkinson and Meg Gelder.
- times: Digital filtering code by Bill Hatch (Coleman Research Corp).
- udina: gnuplot interface to produce beautiful plots from Xlisp-Stat
by Jan de Leeuw and Frederic Udina. kde kernel density estimation
system by Frederic Udina.
- xlisp-to-s: Interface from Xlisp-Stat to something called S or S-plus
or something. By Steve McKinney (U Wash)
- young. Vista: Visual Statistics System by Forrest Young (UNC).
Some of these packages are just snippets, some are major research
efforts. Some have been abandoned in mid-stream, others are still
being developed. I know of quite a few projects that are going on
that we can hopefully add in the near future. By the way, ftp.stat.ucla.edu
is no competitor of statlib, it is simply more specialized and
it can thus be more complete in the areas in which it specializes.
If you have code to contribute, no matter how incomplete, imperfect,
tentative, send it to deleeuw@stat.ucla.edu.
--- Jan