[R-pkgs] New package: logmult (log-multiplicative models)

[Milan Bouchet-Valat]

This is a wrapper around gnm by Turner and Firth to make fitting
log-multiplicative models as convenient as possible: it provides simple
functions, good starting values, jackknife or bootstrap standard errors,
and direct plotting of the results.

In addition, it makes it possible to identify scores from RC(M)
association models, which gnm does not allow without computing the SVD
yourself. It will probably get support for more models in the future.


But let DESCRIPTION speak for itself:

logmult: Log-multiplicative models, including association models

Functions to fit log-multiplicative models using gnm, with support for
convenient printing, plots, and jackknife/bootstrap standard errors.
Currently supported models include UNIDIFF (Erikson & Goldthorpe),
a.k.a. log-multiplicative layer effect model (Xie), and several
association models: Goodman's row-column association models of the RC(M)
and RC(M)-L families resp. for two- and three-way tables with one or
several dimensions; two skew-symmetric association models proposed by
Yamaguchi and by van der Heijden & Mooijaart.


Thanks for your attention. As usual, comments and bug reports are
welcome!



--
Milan Bouchet-Valat
Ph.D. at Observatory for Social Change (OCS-Sciences Po Paris)
Quantitative Sociology Laboratory (LSQ-CREST)
French National Institute for Demographic Studies (INED)