Calculates robust Matthews Correlation Coefficient (MCC), Cohen's Kappa, and robust F-Beta Scores, as introduced by Holzmann and Klar (2026) <doi:10.48550/arXiv.2404.07661>. These performance metrics are designed for imbalanced classification problems. Plots the receiver operating characteristic curve (ROC curve) together with the recall / 1-precision curve.

Robust covariance estimation for matrix-valued data and data with Kronecker-covariance structure using the Matrix Minimum Covariance Determinant (MMCD) estimators and outlier explanation using and Shapley values.

An R package for multiple imputation using chained random forests. Implemented methods can handle missing data in mixed types of variables by using prediction-based or node-based conditional distributions constructed using random forests. For prediction-based imputation, the method based on the empirical distribution of out-of-bag prediction errors of random forests and the method based on normality assumption for prediction errors of random forests are provided for imputing continuous variables. And the method based on predicted probabilities is provided for imputing categorical variables. For node-based imputation, the method based on the conditional distribution formed by the predicting nodes of random forests, and the method based on proximity measures of random forests are provided. More details of the statistical methods can be found in Hong et al. (2020) <arXiv:2004.14823>.

This package provides tools for fitting, diagnosing, and analyzing tie-oriented and actor-oriented relational event models, under both frequentist and Bayesian approaches. The package supports tie-oriented modeling (Butts, 2008, <doi:10.1111/j.1467-9531.2008.00203.x>) and an actor-oriented modeling framework (Stadtfeld et al., 2017, <doi:10.15195/v4.a14>), with additional model diagnostics and goodness-of-fit tools. Interfaces to estimation backends provide a range of extensions: random-effects (frailty) relational event models capturing sender, receiver, and dyadic heterogeneity (Juozaitiene & Wit 2024, <doi:10.1007/s11336-024-09952-x>; Mulder & Hoff, 2024, <doi:10.1214/24-AOAS1885>), finite mixture and dyadic latent class models for unobserved dyadic heterogeneity (Lakdawala et al., 2026, <doi:10.1016/j.socnet.2026.06.006>), penalized estimation via the lasso, ridge, and elastic net (Tibshirani, R., 1996, <doi:10.1111/j.2517-6161.1996.tb02080.x>; Karimova et al., 2023, <doi:10.1016/j.socnet.2023.02.006>), and approximate Bayesian regularization (Karimova et al., 2025, <doi:10.1016/j.jmp.2025.102925>). Modeling of events with a duration is also supported (Lakdawala et al., 2026, <doi:10.48550/arXiv.2602.21000>) and moving window relational event models (Mulder & Leenders, 2019, <doi:10.1016/j.chaos.2018.11.027>; Meijerink et al., 2023, <doi:10.1371/journal.pone.0272309>).

Toolbox that provides a streamlined, end-to-end workflow for propensity score analysis in generating real-world evidence from real-world data. The package covers the full analytic pipeline - from estimating propensity scores via logistic regression, to calculating weights or creating a matched cohort, to generating publication-ready Table 1s with standardized mean differences and weighted balance diagnostics. It also estimates incidence rates, rate differences and rate ratios, hazard ratios, risks, risk ratios, and risk differences, including competing-risk methods and optionally stratified hazard models. Many functions can write formatted Excel reports with method documentation, making results immediately shareable with collaborators and stakeholders. Methods are based on Rosenbaum and Rubin (1983) <doi:10.1093/biomet/70.1.41>, Austin (2011) <doi:10.1080/00273171.2011.568786>, and Desai et al. (2017) <doi:10.1097/EDE.0000000000000595>.

Invoke a BUGS model in OpenBUGS or WinBUGS', a class "bugs" for BUGS results and functions to work with that class. Function write.model() allows a BUGS model file to be written. The class and auxiliary functions could be used with other MCMC programs, including JAGS'. The suggested package BRugs (only needed for function openbugs()) is only available from the CRAN archives, see <https://cran.r-project.org/package=BRugs>.

This package provides an R interface to rclone <https://rclone.org>, a command-line program for managing files on cloud storage. rclone supports over 40 cloud storage providers including S3'-compatible services ('Amazon S3', MinIO', Ceph'), Google Cloud Storage', Azure Blob Storage', and many others. This package downloads and manages the rclone binary automatically and wraps its commands as R functions, returning results as data frames where appropriate.

Supports automated Markov chain Monte Carlo for arbitrarily structured correlation matrices. The user supplies data, a correlation matrix in symbolic form, the current state of the chain, a function that computes the log likelihood, and a list of prior distributions. The package's flagship function then carries out a parameter-at-a-time update of all correlation parameters, and returns the new state. The method is presented in Hughes (2023), in preparation.

This package creates and maintains a build process for complex analytic tasks in R. Package allows to easily generate Makefile for the (GNU) make tool, which drives the build process by (in parallel) executing build commands in order to update results accordingly to given dependencies on changed data or updated source files.

An implementation of an algorithm family for continuous optimization called memetic algorithms with local search chains (MA-LS-Chains), as proposed in Molina et al. (2010) <doi:10.1162/evco.2010.18.1.18102> and Molina et al. (2011) <doi:10.1007/s00500-010-0647-2>. Rmalschains is further discussed in Bergmeir et al. (2016) <doi:10.18637/jss.v075.i04>. Memetic algorithms are hybridizations of genetic algorithms with local search methods. They are especially suited for continuous optimization.

This package provides the first standardized dataset of the Philippines Roll-on/Roll-off (RoRo) shipping network, reflecting the 2024-2026 operational state. It digitizes fragmented records from the Maritime Industry Authority (MARINA) and Philippine Ports Authority (PPA) into a unified framework for transport modeling. The package includes 108 bidirectional provincial links across the Western, Central, and Eastern Nautical Highways, complete with GADM-standardized naming, geospatial coordinates, and metrics such as distance, travel time, and vessel frequency. Methodology follows Anselin (1988, ISBN:9024737354) and LeSage and Pace (2009) <doi:10.1201/9781420064254> for spatial weight construction. Data sources include "MARINA Inventory of RoRo Routes" <https://marina.gov.ph> and "PPA Port Statistics" <https://www.ppa.com.ph/ppa_statistics>. Designed to support research in economic geography and disaster-response logistics.

This package implements the rank-ordered logit (RO-logit) model for stratified analysis of continuous outcomes introduced by Tan et al. (2017) <doi:10.1177/0962280217747309>. Model diagnostics based on the heuristic residuals and estimates in linear scales are available from the package, and outcomes with ties are supported.

Shiny-based interactive gadgets of radial visualization methods and extensions thereof.

R implementation of the common parsing tools lex and yacc'.

Estimation of reproduction numbers for disease outbreak, based on incidence data. The R0 package implements several documented methods. It is therefore possible to compare estimations according to the methods used. Depending on the methods requested by user, basic reproduction number (commonly denoted as R0) or real-time reproduction number (referred to as R(t)) is computed, along with a 95% Confidence Interval. Plotting outputs will give different graphs depending on the methods requested : basic reproductive number estimations will only show the epidemic curve (collected data) and an adjusted model, whereas real-time methods will also show the R(t) variations throughout the outbreak time period. Sensitivity analysis tools are also provided, and allow for investigating effects of varying Generation Time distribution or time window on estimates.

The r4sub package is a meta-package that installs and loads core packages of the R4SUB (R for Regulatory Submission) clinical submission readiness ecosystem. Loading r4sub attaches r4subcore', r4subtrace', r4subscore', r4subrisk', r4subdata', and r4subprofile'.

An interface to the Mangal database - a collection of ecological networks. This package includes functions to work with the Mangal RESTful API methods (<https://mangal-interactions.github.io/mangal-api/>).

Exploration of pharmacometrics data involves both general tools (transformation and plotting) and specific techniques (non-compartmental analysis). This kind of exploration is generally accomplished by utilizing different packages. The purpose of ruminate is to create a shiny interface to make these tools more broadly available while creating reproducible results.

This package provides methods from Yeh, Rice, and Dubin (2022) <doi:10.1080/00031305.2021.1967781> for comparing two continuously updated probabilistic forecasts under squared (Brier) loss: pointwise loss and variance, a global delta test (Monte Carlo p-values), simulation designs, and a naive pointwise band plot.

Allows for production of Czekanowski's Diagrams with clusters. See K. Bartoszek, A. Vasterlund (2020) <doi:10.2478/bile-2020-0008> and K. Bartoszek, Y. Luo (2023) <doi:10.14708/ma.v51i2.7259>. The suggested FuzzyDBScan package (which allows for fuzzy clustering) can be obtained from <https://github.com/henrifnk/FuzzyDBScan/> (or from CRAN's Archive <https://cran.r-project.org/src/contrib/Archive/FuzzyDBScan/>).

Integrated tools to support rigorous and well documented data harmonization based on Maelstrom Research guidelines. The package includes functions to assess and prepare input elements, apply specified processing rules to generate harmonized datasets, validate data processing and identify processing errors, and document and summarize harmonized outputs. The harmonization process is defined and structured by two key user-generated documents: the DataSchema (specifying the list of harmonized variables to generate across datasets) and the Data Processing Elements (specifying the input elements and processing algorithms to generate harmonized variables in DataSchema formats). The package was developed to address key challenges of retrospective data harmonization in epidemiology (as described in Fortier I and al. (2017) <doi:10.1093/ije/dyw075>) but can be used for any data harmonization initiative.

The TRUST4 or MiXCR is used to identify the clonotypes. The goal of rTCRBCRr is to process the results from these clonotyping tools, and analyze the clonotype repertoire metrics based on chain names and IGH isotypes. The manuscript is still under preparation for publication for now. The references describing the methods in this package will be added later.

This package provides functions for estimating sparse precision matrices using a random graphical model framework under an L0-style penalty. The method evaluates candidate theta values and returns both continuous and binary precision matrices representing inferred network structures.

With this package we provide an easy method to compute robust and conditional Data Envelopment Analysis (DEA), Free Disposal Hull (FDH) and Benefit of the Doubt (BOD) scores. The robust approach is based on the work of Cazals, Florens and Simar (2002) <doi:10.1016/S0304-4076(01)00080-X>. The conditional approach is based on Daraio and Simar (2007) <doi:10.1007/s11123-007-0049-3>. Besides we provide graphs to help with the choice of m. We relay on the Benchmarking package to compute the efficiency scores and on the np package to compute non parametric estimation of similarity among units.