Tools for the ‘Parallel’ Package (partools)
Miscellaneous utilities for the ‘parallel’ package, cluster-oriented section; ‘Snowdoop’ alternative to MapReduce; file splitting and distributed operations such as sort and aggregate. ‘Software Alchemy’ method for parallelizing most statistical methods, presented in N. Matloff, Parallel Computation for Data Science, Chapman and Hall, 2015. Includes a debugging aid.
Weighted Effect Coding (wec)
Provides functions to create factor variables with contrasts based on weighted effect coding. In weighted effect coding the estimates from a first order regression model show the deviations per group from the sample mean. This is especially useful when a researcher has no directional hypotheses and uses a sample from a population in which the number of observation per group is different The package also provides functionality for interactions between two factor variables based on weighted effect coding. Please note that this is a beta version: while functional, it does not follow all R conventions.
Threshold Regression (threg)
Fit a threshold regression model based on the first-hitting-time of a boundary by the sample path of a Wiener diffusion process. The threshold regression methodology is well suited to applications involving survival and time-to-event data.
Lasso Penalization in Large Scale Generalized Linear Array Models (glamlasso)
Efficient design matrix free procedure for Lasso regularized estimation in large scale 3-dimensional generalized linear array models. The Gaussian model with identity link, the Binomial model with logit link, the Poisson model with log link and the Gamma model with log link is currently implemented.