This package provides a set of basic and extensible data structures and functions for multivariate analysis, including dimensionality reduction techniques, projection methods, and preprocessing functions. The aim of this package is to offer a flexible and user-friendly framework for multivariate analysis that can be easily extended for custom requirements and specific data analysis tasks.

Estimation and bootstrap utilities for single layer and multilayer Mixed Graphical Models, including functions for centrality, bridge metrics, membership stability, and plotting (De Martino et al. (2026) <doi:10.48550/arXiv.2602.05716>).

This package provides functions for multivariate and propensity score matching and for finding optimal balance based on a genetic search algorithm. A variety of univariate and multivariate metrics to determine if balance has been obtained are also provided. For details, see the paper by Jasjeet Sekhon (2007, <doi:10.18637/jss.v042.i07>).

Estimate genetic linkage maps for markers on a single chromosome (or in a single linkage group) from pairwise recombination fractions or intermarker distances using weighted metric multidimensional scaling. The methods are suitable for autotetraploid as well as diploid populations. Options for assessing the fit to a known map are also provided. Methods are discussed in detail in Preedy and Hackett (2016) <doi:10.1007/s00122-016-2761-8>.

Sentiment analysis is a popular technique in text mining that attempts to determine the emotional state of some text. We provide a new implementation of a common method for computing sentiment, whereby words are scored as positive or negative according to a dictionary lookup. Then the sum of those scores is returned for the document. We use the Hu and Liu sentiment dictionary ('Hu and Liu', 2004) <doi:10.1145/1014052.1014073> for determining sentiment. The scoring function is vectorized by document, and scores for multiple documents are computed in parallel via OpenMP'.

Loads, validates, and replays portable mars ModelSpec artifacts from R. The package provides helpers for constructing design matrices, generating predictions, and, when the companion runtime helper is available, fitting portable model specifications for cross-language replay.

This package provides a web-based graphical user interface to provide the basic steps of a machine learning workflow. It uses the functionalities of the mlr3 framework.

The Iterative Proportional Fitting (IPF) algorithm operates on count data. This package offers implementations for several algorithms that extend this to nested structures: parent and child items for both of which constraints can be provided. The fitting algorithms include Iterative Proportional Updating <https://trid.trb.org/view/881554>, Hierarchical IPF <doi:10.3929/ethz-a-006620748>, Entropy Optimization <https://trid.trb.org/view/881144>, and Generalized Raking <doi:10.2307/2290793>. Additionally, a number of replication methods is also provided such as Truncate, replicate, sample <doi:10.1016/j.compenvurbsys.2013.03.004>.

Estimation of multivariate differences between two groups (e.g., multivariate sex differences) with regularized regression methods and predictive approach. See Ilmarinen et al. (2023) <doi:10.1177/08902070221088155>. Deconstructing difference score correlations (e.g., gender-equality paradox), see Ilmarinen & Lönnqvist (2024) <doi:10.1037/pspp0000508>. Includes also tools that help in understanding difference score reliability, conditional intra-class correlations, tail-dependency, and heterogeneity of variance estimates. Package development was supported by the Academy of Finland research grant 338891.

Interface to Apache Commons Email to send emails from R.

This package provides tools for motif analysis in multi-level networks. Multi-level networks combine multiple networks in one, e.g. social-ecological networks. Motifs are small configurations of nodes and edges (subgraphs) occurring in networks. motifr can visualize multi-level networks, count multi-level network motifs and compare motif occurrences to baseline models. It also identifies contributions of existing or potential edges to motifs to find critical or missing edges. The package is in many parts an R wrapper for the excellent SESMotifAnalyser Python package written by Tim Seppelt.

Acoustic template detection and monitoring database interface. Create, modify, save, and use templates for detection of animal vocalizations. View, verify, and extract results. Upload a MySQL schema to a existing instance, manage survey metadata, write and read templates and detections locally or to the database.

This package provides a simple way to construct and maintain functions that keep state i.e. remember their argument lists. This can be useful when one needs to repeatedly invoke the same function with only a small number of argument changes at each invocation.

Detection of multivariate outliers using robust estimates of location and scale. The Minimum Covariance Determinant (MCD) estimator is used to calculate robust estimates of the mean vector and covariance matrix. Outliers are determined based on robust Mahalanobis distances using either an unstructured covariance matrix, a principal components structured covariance matrix, or a factor analysis structured covariance matrix. Includes options for specifying the direction of interest for outlier detection for each variable.

Get map data frames for the Indian subcontinent with different region levels (e.g., district, state). The package also offers convenience functions for plotting choropleths, visualizing spatial data, and handling state/district codes.

Multiple Imputation has been shown to be a flexible method to impute missing values by Van Buuren (2007) <doi:10.1177/0962280206074463>. Expanding on this, random forests have been shown to be an accurate model by Stekhoven and Buhlmann <arXiv:1105.0828> to impute missing values in datasets. They have the added benefits of returning out of bag error and variable importance estimates, as well as being simple to run in parallel.

An extensible expectation-maximization (EM) framework for finite mixtures of quantile regressions (clusterwise / mixture-of-experts quantile regression). A single EM substrate with an engine/extension contract carries a family of capabilities: the core free-weight mixture of Wu and Yao (2016) <doi:10.1016/j.csda.2014.04.014> -- a fast asymmetric-Laplace path and the nonparametric kernel-density EM with components constrained to have their tau-quantile equal to zero (Hall and Presnell 1999 device); expectile and M-quantile component-loss families (Newey and Powell 1987; Breckling and Chambers 1988); component-specific penalized variable selection (SCAD / adaptive-LASSO, the quantile analogue of Khalili and Chen 2007); and joint multi-quantile estimation with a shared latent classification and non-crossing component curves. Provides classification-aware standard errors (sparsity and stochastic-EM multiple imputation), multi-start estimation, component-count selection, and prediction. The companion package mixqrgate adds location-varying gating.

This package provides functions and datasets to support Smilde, Næs and Liland (2021, ISBN: 978-1-119-60096-1) "Multiblock Data Fusion in Statistics and Machine Learning - Applications in the Natural and Life Sciences". This implements and imports a large collection of methods for multiblock data analysis with common interfaces, result- and plotting functions, several real data sets and six vignettes covering a range different applications.

This package provides tools and functions to fit a multilevel index of dissimilarity.

Allows to perform the multivariate version of the Diebold-Mariano test for equal predictive ability of multiple forecast comparison. Main reference: Mariano, R.S., Preve, D. (2012) <doi:10.1016/j.jeconom.2012.01.014>.

The latest guidelines proposed by International Expert Consensus are used for the clinical diagnosis of Metabolic Associated Fatty Liver Disease (MAFLD). The new definition takes hepatic steatosis (determined by elastography or histology or biomarker-based fatty liver index) as a major criterion. In addition, race, gender, body mass index (BMI), waist circumference (WC), fasting plasma glucose (FPG), systolic blood pressure (SBP), diastolic blood pressure (DBP), triglycerides (TG), high-density lipoprotein cholesterol (HDLC), homeostatic model assessment of insulin resistance (HOMAIR), high sensitive c-reactive protein (HsCRP) for the diagnosis of MAFLD. Each parameter has to be interpreted based on the proposed cut-offs, making the diagnosis slightly complex and error-prone. This package is developed by incorporating the latest international expert consensus guidelines, and it will aid in the easy and quick diagnosis of MAFLD based on FibroScan in busy healthcare settings and also for research purposes. The new definition for MAFLD as per the International Consensus Statement is described by Eslam M et al (2020). <doi:10.1016/j.jhep.2020.03.039>.

Network meta-analysis and network meta-regression models for aggregate data, individual patient data, and mixtures of both individual and aggregate data using multilevel network meta-regression as described by Phillippo et al. (2020) <doi:10.1111/rssa.12579>. Models are estimated in a Bayesian framework using Stan'.

This package provides classes to implement, analyze and plot cohort life tables for actuarial calculations. Birth-year dependent cohort mortality tables using a yearly trend to extrapolate from a base year are implemented, as well as period life table, cohort life tables using an age shift, and merged life tables. Additionally, several data sets from various countries are included to provide widely-used tables out of the box.

This package implements the Multi-view Aggregated Two-Sample (MATES) test, a powerful nonparametric method for testing equality of two multivariate distributions. The method constructs multiple graph-based statistics from various perspectives (views) including different distance metrics, graph types (nearest neighbor graphs, minimum spanning trees, and robust nearest neighbor graphs), and weighting schemes. These statistics are then aggregated through a quadratic form to achieve improved statistical power. The package provides both asymptotic closed-form inference and permutation-based testing procedures. For methodological details, see Cai and others (2026+) <doi:10.48550/arXiv.2412.16684>.