Plackett-Luce models estimate the underlying worth of items compared in a set of rankings, e.g. the ability of athletes given a set of race results, the attractiveness of perfumes given a set of consumer preference rankings. This package accomodates ties in the rankings as well as partial rankings (only ranking a subset of the items each time).
For more details, see the PlackettLuce webpage.
These models are like generalized linear models (linear regression, logistic regression, log-linear models, etc.) but may also include one or more nonlinear terms in the predictor function (i.e., the right hand side of the regression equation).
For an introduction, see the R News article (for regular R users) or the Statistical Computing & Graphics Newsletter article (for those less familiar with R). The Talks page has slides from various talks related to the package, covering both technical details and applications. More information, including workshop slides, is available on the gnm webpage.
Bradley-Terry models are for modelling pair comparison data, e.g. to predict the winner in a paired contest based on player ability. This package covers models in which the ability or worth is modelled by covariates.
Generalized Semi-linear Canonical Correlation Analysis (GSLCCA) estimates the parameters of a given nonlinear model to maximize the correlation with a linear combination of multiple response variables. This method was developed to characterize EEG power spectra under different treatment regimens in research and development projects at Pfizer.
The package is still in development. Further details, including the current development version can be found via the gslcca project homepage on R Forge.
vcdExtra: Extra methods for visualising categorical data
This package complements the vcd package for visualising categorical data, providing additional methods and examples.
See the vcdExtra webpage on R-Forge for more information.