development | CRAN

, available onenmSdmXCRAN, is a set of tools in R for implementing species distribution models (SDMs) and ecological niche models (ENMs), including: bias correction, spatial cross-validation, model evaluation, raster interpolation, biotic “velocity” (speed and direction of movement of a “mass” represented by a raster), and tools for using spatially imprecise records. The heart of the package is a set of “training” functions which automatically optimize model complexity based number of available occurrences. These algorithms include MaxEnt, MaxNet, boosted regression trees/gradient boosting machines (BRT), generalized additive models (GAM), generalized linear models (GLM), natural splines, and random forests (RF). To enhance interoperability with other packages, the package does not create any new classes. The package works with PROJ6 geodetic objects and coordinate reference systems.

R package

uses the stand-alone installer of Open Source Geospatial’s GRASS GIS Version 8 to speed up some commonly used raster and vector operations. Most of these operations can be done using thefasterRasterrasteror newerterrapackages by Robert Hijmans, or thergeosor newersfpackages. However, when the input raster or vector is very large in memory, in some cases functions in those packages can take a long time and fail. ThefasterRasterpackage attempts to address these problems by calls to GRASS which is faster. Please note thatterraandsfmay be faster and thus the better solution for functions that this package implements. However, in some casesfasterRasteris still faster!

R package * statisfactory*, available on CRAN, is a warehouse of statistical tools and helper functions for back-transforming of principal component scores, creation all possible formulae from a set of terms while respecting marginality, stratified sampling, and 2-D histograms, amongst others.

R package * mcmcHammer *is just like all the other packages for analyzing MCMC chains from Bayesian analysis, except that it’s not. Like the others, it can create trace plots and density plots. But unlike them, it automates “extracting” variables, especially variables with indices. For example, say your set of MCMC chains have variables named

*beta0*,

*beta1*, and

*beta2*, as well as

*gamma[1, 1]*,

*gamma[1, 2]*,

*gamma[2, 1]*, and

*gamma[2, 2]*. You can easily create trace plots and density plots for each of these with minimal “manual” tweaking of variable names.

R package * omnibus*, on CRAN: An assortment of helper functions for managing data (e.g., rotating values in matrices by a user-defined angle, switching from row- to column-indexing), dates (e.g., intuiting year from messy date strings), handling missing values (e.g., removing elements/rows across multiple vectors or matrices if any have an NA), and text (e.g., flushing reports to the console in real-time).