I’m involved in the following R packages which are all available on CRAN and which can be directly installed in R after choosing a mirror using the command:
install.packages(“pkg_name”,dependencies=”Depends”)
Blind source separation for multivariate spatio-temporal data
Simultaneous/joint diagonalization of local autocovariance matrices to estimate spatio-temporally uncorrelated random fields.
Stationary subspace analysis
Stationary subspace analysis (SSA) is a blind source separation (BSS) variant where stationary components are separated from non-stationary components. Several SSA methods for multivariate time series are provided here along with functions to simulate time series with time-varying variance and autocovariance.
M-Estimation of Shape for Data with Missing Values
M-estimators of location and shape following the power family are provided in the case of complete data and also when observations have missing values together with functions aiding their visualization.
Whitening Data as Preparation for Blind Source Separation
Whitening is the first step of almost all blind source separation (BSS) methods. A fast implementation of whitening for BSS is implemented to serve as a lightweight dependency for packages providing BSS methods.
Kernel Independent Component Analysis
The kernel independent component analysis (kernel ICA) method as introduced by Bach and Jordan (2003). The incomplete Cholesky decomposition used in kernel ICA is provided as separate function.
‘REPPlab’ Via a Shiny Application.
Performs exploratory projection pursuit via ‘REPPlab’ using a Shiny app.
Blind Source Separation for Multivariate Spatial Data
Perform blind source separation for multivariate spatial data based on simultaneous/joint diagonalization of local covariance matrices.
ICS via a Shiny Application
Performs Invariant Coordinate Selection (ICS) and especially ICS outlier identification using a shiny app.
Outlier Detection Using Invariant Coordinate Selection
Multivariate outlier detection is performed using invariant coordinates where the package offers different methods to choose the appropriate components.
Estimating and Testing the Number of Interesting Components in Linear Dimension Reduction
For different linear dimension reduction methods like principal components analysis (PCA), independent components analysis (ICA) and supervised linear dimension reduction tests and estimates for the number of interesting components (ICs) are provided.
Blind Source Separation Methods for Tensor-Valued Observations
Contains several utility functions for manipulating tensor-valued data (centering, multiplication from a single mode etc.) and the implementations of the following blind source separation methods for tensor-valued data: ‘tPCA’, ‘tFOBI’, ‘tJADE’, ‘tgFOBI’, ‘tgJADE’ and ‘tSOBI’.
Tools for Blind Source Separation for Time Series
Different estimates are provided to solve the blind source separation problem for time series with stochastic volatility.
Asymptotic Covariance Matrices of Some BSS Mixing and Unmixing Matrix Estimates
Functions to compute the asymptotic covariance matrices of mixing and unmixing matrix estimates of the following blind source separation (BSS) methods: symmetric and squared symmetric FastICA, regular and adaptive deflation-based FastICA, FOBI, JADE, AMUSE and deflation-based and symmetric SOBI. Also functions to estimate these covariances based on data are available.
Classical, Reloaded and Adaptive FastICA Algorithms
Algorithms for classical symmetric and deflation-based FastICA, reloaded deflation-based FastICA algorithm and an algorithm for adaptive deflation-based FastICA using multiple nonlinearities.
Blind Source Separation Methods Based on Joint Diagonalization and Some BSS Performance Criteria
Cardoso’s JADE algorithm as well as his functions for joint diagonalization are ported to R. Also several other blind source separation (BSS) methods, like AMUSE and SOBI, and some criteria for performance evaluation of BSS algorithms, are given.
Fast Computation of Multivariate M-estimators
The package implements the new algorithm for fast computation of M-scatter matrices using a partial Newton-Raphson procedure for several estimators.
The new algorithms are described in detail in Dümbgen, L., Nordhausen, K. and Schuhmacher, H. (2016): New Algorithms for M-estimation of Multivariate Location and Scatter. Journal of Multivariate Analysis, 144, 200-217.
Tools for Linear Dimension Reduction
Linear dimension reduction subspaces can be uniquely defined using orthogonal projection matrices. This package provides tools to compute distances between such subspaces and to compute the average subspace.
R Interface to ‘EPP-Lab’, a Java Program for Exploratory Projection Pursuit
An R Interface to ‘EPP-lab’ v1.0. ‘EPP-lab’ is a Java program for projection pursuit using genetic algorithms written by Alain Berro and S. Larabi Marie-Sainte and is included in the package. The ‘EPP-lab’ sources are available under https://github.com/fischuu/EPP-lab.git.
Multivariate Methods Based on the Oja Median and Related Concepts
Functions to calculate the Oja median, Oja signs and ranks and methods based upon them.
Multivariate Nonparametric Methods. An Approach Based on Spatial Signs and Ranks
Multivariate tests, estimates and methods based on the identity score, spatial sign score and spatial rank score are provided. The methods include one and c-sample problems, shape estimation and testing, linear regression and principal components.
The package accompanies the book Oja (2010) and is in detailed described in Nordhausen, K. and Oja, H. (2011): Multivariate L1 Methods: The package MNM. Journal of Statistical Software, 43, 1-28.
Multivariate nonparametric methods based on spatial signs and ranks
This package contains test and estimates of location, tests of independence, tests of sphericity and several estimates of shape all based on spatial signs, symmetrized signs, ranks and signed ranks.
Tools for Multivariate Nonparametrics
Tools for multivariate nonparametrics, as location tests based on marginal ranks, spatial median and spatial signs computation, Hotelling’s T-test, estimates of shape are implemented.
Tools for Exploring Multivariate Data via ICS/ICA
Implementation of Tyler, D.E., Critchley, F., Dümbgen, L., and Oja, H. (2009) and Oja, H., Sirkiä, S. and Eriksson, J. (2006) method of two different scatter matrices to obtain an invariant coordinate system or independent components, depending on the underlying assumptions.
Described in detail in Nordhausen, K., Oja, H. and Tyler, D.E. (2008): Tools for Exploring Multivariate Data: The Package ICS. Journal of Statistical Software, 28, 1-31.
Any feedback, suggestions and bug reports are welcome! Just send an email!