Over the last couple of years I have written quite a few
R programs for various "psychometric" techniques, and I am
regularly updating and expanding what is there. I now
have (wholly or partially), or have planned
-- gifi package (update to homals on CRAN). Code for
multiple correspondence analysis, nonlinear principal
component analysis, nonlinear multiset canonical
correlation analysis. Implements everything in Gifi (1990)
or SPSS Categories, and then some. Status: done.
-- aspect package (optimizing functions of correlation
matrices over transformations/quantifications of the variables --
functions implemented are sums of eigenvalues, determinants,
multiple
correlations, power sums). Status: done.
-- optim package (miscellaneous optimization routines -- currently has
pooled-adjacent-violaters for monotone regression and
the hildreth-desopo coordinate descent method for quadratic
programming). Status: will grow.
-- ca package (simple correspondence analysis with various
plotting options). Status: done.
-- smacof package (metric and nonmetric multidimensional
scaling, individual difference models, rectangular matrices,
constrained scaling, metric nearness problem, Shepard-Luce
models). Status: mostly done.
-- logitfold package (logistic likelihood solution to nonmetric
unfolding for binary and multicategory data, multidimensional
choice models, logistic principal component analysis and
canonical analysis, roll-call models, multidimensional Rasch
models).
Status: partly done.
-- lssem package (linear structural equation models with latent
variables, reformulated as matrix decomposition problems, and
solved by majorizing least squares loss functions). Status:
planning stage.
-- threeway package (multiway generalizations of principal component
analysis with various constraints on the decomposition). Status:
on the horizon.
All of this is currently in straightforward R, without any compiled C
code,
and without any OOP.
The idea is that eventually these will be nicely organized in
R packages that pass the checks and are internally documented.
But "eventually" can take a pretty long time. Also, I expect that
some of this will wind up on www.jstatsoft.org, but again,
probably not any time soon.
Also, I would be more than willing to embed this stuff in general
multi-person projects such as
R in Psychometrics
R in Econometrics
R in Social Statistics
but that will require someone else running these shows.
For the time being, if you want to receive information about updates
of this
and related R code and papers in the area of multidimensional
scaling, item response theory, choice models, factor analysis,
and simultaneous equation models, please subscribe to
http://lists.stat.ucla.edu/mailman/listinfo/albertgifi
==Jan de Leeuw; Distinguished Professor and Chair, UCLA Department of
Statistics;
Editor: Journal of Multivariate Analysis, Journal of Statistical
Software
US mail: 8130 Math Sciences Bldg, Box 951554, Los Angeles, CA 90095-1554
phone (310)-825-9550; fax (310)-206-5658; email: deleeuw at stat.ucla.edu
.mac: jdeleeuw ++++++ aim: deleeuwjan ++++++ skype: j_deleeuw
homepages: http://gifi.stat.ucla.edu ++++++ http://www.cuddyvalley.org
------------------------------------------------------------------------
-------------------------
No matter where you go, there you are. --- Buckaroo Banzai
http://gifi.stat.ucla.edu/sounds/nomatter.au
