Provides methods for detecting multiple change-points and segmenting univariate time series using Smallest Valid Partitioning (SVP). SVP searches for a partition with the smallest number of segments whose segments satisfy a user-defined or built-in validity test. Among partitions with the same number of segments, it minimizes a within-segment sum-of-squared-errors criterion.

Enables high dimensional statistical modeling using Bayesian inference and provides a probabilistic programming language for high dimensional problems. See Kleinsasser (2026) &lt;<a href="https://doi.org/10.5281%2Fzenodo.22309216" target="_top">doi:10.5281/zenodo.22309216</a>&gt;.

Implements partially linear panel regression (PLPR) models with interactive fixed effects, high-dimensional confounding variables, and an exogenous treatment variable within the double machine learning framework. Estimates the structural parameter (treatment effect) in static panel data models with interactive fixed effects using the approach established in Chen et al. (2026) &lt;<a href="https://doi.org/10.48550%2FarXiv.2608.01137" target="_top">doi:10.48550/arXiv.2608.01137</a>&gt;. Builds on the object-oriented package 'DoubleML' (Bach et al., 2024) &lt;<a href="https://doi.org/10.18637%2Fjss.v108.i03" target="_top">doi:10.18637/jss.v108.i03</a>&gt; and 'xtdml' (Polselli, 2025) &lt;<a href="https://doi.org/10.48550%2FarXiv.2512.15965" target="_top">doi:10.48550/arXiv.2512.15965</a>&gt;, using the 'mlr3' ecosystem.

Provides tools for identifying potential information leakage and validation risks in machine learning workflows using environmental sensor data. The package includes checks for temporal ordering, shared sensors, spatial proximity, and overlapping temporal windows. The diagnostics are motivated by considerations of spatial and temporal structure in model validation (Roberts et al., 2017) &lt;<a href="https://doi.org/10.1111%2Fecog.02881" target="_top">doi:10.1111/ecog.02881</a>&gt;.

The original implementation of the Forward Angular Relevance Measure (FARM). The algorithm relies on a forward alignment approach for compared time series based on the dynamic time warping (DTW) principle. It considers the differences between data point as source in a combined distance metric that is used for series alignment. The algorithm returns a global and a series of local relevance measures relying on correlation coefficients. A normalization of time series is not part of the algorithm but recommended for best results. The FARM method is introduced in: Christen et al. (2023) &lt;<a href="https://doi.org/10.48550%2FarXiv.2304.11028" target="_top">doi:10.48550/arXiv.2304.11028</a>&gt;.

Download, read and tidy the ESTBAN (Estatistica Bancaria Mensal por Municipio, Monthly Banking Statistics by Municipality) files published by the Brazilian Central Bank (Banco Central do Brasil) for every bank branch and municipality in Brazil. Each file reports balance-sheet accounts of the COSIF (Plano Contabil das Instituicoes do Sistema Financeiro Nacional, the chart of accounts of the Brazilian financial system) such as credit operations, deposits and savings. Files are fetched from the official site &lt;<a href="https://www.bcb.gov.br/estatisticas/estatisticabancariamunicipios" target="_top">https://www.bcb.gov.br/estatisticas/estatisticabancariamunicipios</a>&gt; with an idempotent local cache, read from their Latin-1 encoded CSV (comma-separated values) layout into tibbles, optionally filtered by state, and aggregated by municipality. Includes tools to detect and impute institution-month non-reports (an institution present in the file with every account equal to zero), which would otherwise be mistaken for zero balances.

Fits and compares generalized linear mixed models for multiple ecological responses and environmental predictors. The package supports additive and temporal-interaction candidate models, AICc model selection, likelihood-ratio tests, coefficient extraction, Nakagawa R-squared, simulation-based diagnostics, figures, and spreadsheet exports. Model selection follows Burnham and Anderson (2002, ISBN:9780387953649); marginal and conditional R-squared follow Nakagawa and Schielzeth (2013) &lt;<a href="https://doi.org/10.1111%2Fj.2041-210x.2012.00261.x" target="_top">doi:10.1111/j.2041-210x.2012.00261.x</a>&gt;.

Download and read the public, non-identified microdata of the Novo CAGED (Cadastro Geral de Empregados e Desempregados), the monthly registry of formal employment movements published by the Brazilian Ministry of Labour and Employment through the PDET FTP server &lt;<a href="ftp://ftp.mtps.gov.br/pdet/microdados/" target="_top">ftp://ftp.mtps.gov.br/pdet/microdados/</a>&gt;. Lists the reference months available on the server, downloads the three monthly files (movements declared on time, declared late, and exclusions) with an idempotent local cache, and reads the national 7z archives as a stream, filtering by state and selecting columns before anything is kept in memory, so that a single state can be extracted without loading the full national file. Also provides the official record layout and a helper to consolidate admissions, separations and net balance by reference month.

Acquire and audit public police stop and search archives for England and Wales, retaining provenance and explicit coverage information. Designed for exposure-based ethnic disparity analysis with separately reported sampling and assumption uncertainty. Contains public sector information licensed under the Open Government Licence v3.0.

Provides validation, harmonization, descriptive analysis, compositional analysis, transparent risk components, grouped cross-validation, visualization, and predictive modelling tools for freshwater microplastic datasets. The package includes synthetic demonstration data conforming to the LIMPID-India data model and emphasizes explicit units, provenance, percentage closure, non-imputation of missing environmental covariates, and leakage-aware model evaluation.

Fits area-level geoadditive small area estimation (SAE) models by extending the Fay-Herriot area-level model with linear, nonlinear, and spatial effects. The Fay-Herriot model is described by Fay and Herriot (1979) &lt;<a href="https://doi.org/10.1080%2F01621459.1979.10482505" target="_top">doi:10.1080/01621459.1979.10482505</a>&gt;. Geoadditive models combine nonlinear covariate effects and spatial variation as described by Kammann and Wand (2003) &lt;<a href="https://doi.org/10.1111%2F1467-9876.00385" target="_top">doi:10.1111/1467-9876.00385</a>&gt;, while their application to small area estimation is discussed by Pusponegoro et al. (2019) &lt;<a href="https://doi.org/10.21108%2FJDSA.2019.2.15" target="_top">doi:10.21108/JDSA.2019.2.15</a>&gt;. Nonlinear covariate effects are represented using penalized splines, while spatial effects are represented using a smooth function of geographic coordinates. Models are estimated using restricted maximum likelihood (REML), and mean squared error (MSE) is estimated using a parametric bootstrap. The package also provides comparisons with the Fay-Herriot and spatial Fay-Herriot (SFH) models.

Provides an interactive Shiny application and a toolbox of R functions for the management, calculation, filtering, visualization and exploratory analysis of molecular descriptors and ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) properties of small molecules. Computes descriptors locally via the Chemistry Development Kit (CDK), and offers drug-likeness filters (Lipinski, Veber, Ghose, Egan, Muegge), the BOILED-Egg model for gastrointestinal absorption and blood-brain barrier permeability, a P-glycoprotein (P-gp, also known as ATP-binding cassette sub-family B member 1, ABCB1) substrate Random Forest classifier, Principal Component Analysis (PCA), t-Distributed Stochastic Neighbor Embedding (t-SNE), Uniform Manifold Approximation and Projection (UMAP), radar plots and Tanimoto / AGglomerative NESting (AGNES) clustering to support compound prioritization in early-stage drug discovery.

Imputes missing biomarker measurements in a wide longitudinal serum panel with gradient-boosted decision trees, groups the completed panel by Bayesian consensus clustering, and compares the resulting patient subgroups by Kaplan-Meier, log-rank and Cox analysis. The imputation learner is described in Ke et al. (2017) &lt;<a href="https://papers.nips.cc/paper/6907-lightgbm-a-highly-efficient-gradient-boosting-decision-tree" target="_top">https://papers.nips.cc/paper/6907-lightgbm-a-highly-efficient-gradient-boosting-decision-tree</a>&gt; and the clustering method in Lock and Dunson (2013) &lt;<a href="https://doi.org/10.1093%2Fbioinformatics%2Fbtt425" target="_top">doi:10.1093/bioinformatics/btt425</a>&gt;. Imputed values are conditional-mean predictions, so the procedure is a machine-learning single imputation; the completions carry no between-imputation variance and must not be pooled by Rubin's rules. Two panels from Gene Expression Omnibus accession 'GSE65622' are included, one for each survival endpoint.

Provides short and consistent commands for data management, descriptive and inferential statistics, epidemiological analyses, regression models, survival and longitudinal analyses, diagnostic accuracy, scale assessment, meta-analysis, machine learning, study design, publication-ready tables, graphics, and reporting. Commands accept an explicit data frame or an active data frame selected with usedf().

Scalable implementation of the Wilcoxon rank sum test and the area under the receiver operating characteristic curve (also known as auROC or AUC) statistic. Interfaces to dense and sparse matrices, as well as the genomics analysis frameworks 'Seurat' and 'SingleCellExperiment'.

Provides tools to compute individual percentile ranks and z-scores for clinical biomarkers in children, adolescents and young adults, based on age-, sex-, and height-specific reference data from the IDEFICS (Identification and prevention of Dietary and lifestyle-induced health EFfects In Children and infantS) study and the Biomarkers4Pediatrics collaboration. Supports the computation of a composite Metabolic Syndrome (MetS) score and associated monitoring/action levels for health monitoring. For more details see Ahrens et al. (2014) &lt;<a href="https://doi.org/10.1038%2Fijo.2014.130" target="_top">doi:10.1038/ijo.2014.130</a>&gt;.

Provides user-friendly methods for handling missing data in regression modeling, including available-case linear regression, multiple imputation, random-forest imputation, and the Tower method. Implemented approaches include chained-equation imputation described by van Buuren and Groothuis-Oudshoorn (2011) &lt;<a href="https://doi.org/10.18637%2Fjss.v045.i03" target="_top">doi:10.18637/jss.v045.i03</a>&gt;, multiple imputation described by Honaker, King and Blackwell (2011) &lt;<a href="https://doi.org/10.18637%2Fjss.v045.i07" target="_top">doi:10.18637/jss.v045.i07</a>&gt;, random-forest imputation described by Stekhoven and Buehlmann (2012) &lt;<a href="https://doi.org/10.1093%2Fbioinformatics%2Fbtr597" target="_top">doi:10.1093/bioinformatics/btr597</a>&gt;, and the Tower method described by Matloff and Mohanty (2023) &lt;<a href="https://CRAN.R-project.org/package=toweranNA" target="_top">https://CRAN.R-project.org/package=toweranNA</a>&gt;.

Provides a single-producer single-consumer channel with bounded capacity, optional timeouts on receives, and sentinel return values for full, timeout, and closed states. The ring buffer is implemented in C: sends and receives are constant-time operations, values are delivered in first-in first-out order, and a closed channel drains its buffered values before reporting the closed state. The transport is in-process: both ends of a channel live in the calling R process, which suits the channel to prototyping producer and consumer designs and to testing channel-based logic.

Implements a minimal 'LaTeX' parser that translates mathematical formulas into R code strings. Supports arithmetic operators, implicit multiplication, fractions, Greek letters, common mathematical functions, and statistical notation for means, medians, and rolling sums. Input from visual formula editors such as 'MathQuill' is normalized automatically, and the resulting string can be evaluated with parse() and eval(), or converted into an R function with latex2fun(). The implementation follows the tree-walking interpreter design of Nystrom (2021) &lt;<a href="https://craftinginterpreters.com/" target="_top">https://craftinginterpreters.com/</a>&gt;.

Provides Korean morphological analysis, part-of-speech (POS) tagging, noun extraction, Hangul (Korean script) conversion utilities, concordance search, mutual information statistics, and user dictionary management tools for Korean text research. The morphological analyzer backend is derived from the 'KoNLP' package and the 'HanNanum' analyzer, and is reimplemented in native C so that no Java runtime is required. Bundled dictionary and statistical resources allow analysis to work out of the box, and optional dictionaries can be loaded at runtime from the 'Sejong' and 'NIADic' packages when they are installed and loaded.

Provides high-performance implementations of Small Area Estimation (SAE) methods leveraging C++ ('Rcpp', 'RcppArmadillo') and 'OpenMP' multi-threading. Supports standard area-level Fay-Herriot models (Fay and Herriot, 1979 &lt;<a href="https://doi.org/10.1080%2F01621459.1979.10482505" target="_top">doi:10.1080/01621459.1979.10482505</a>&gt;), Spatial Fay-Herriot models (Pratesi and Salvati, 2008 &lt;<a href="https://doi.org/10.1002%2Fenv.861" target="_top">doi:10.1002/env.861</a>&gt;), Spatio-Temporal Fay-Herriot models (Marhuenda et al., 2013 &lt;<a href="https://doi.org/10.1016%2Fj.csda.2013.01.016" target="_top">doi:10.1016/j.csda.2013.01.016</a>&gt;), and unit-level Battese-Harter-Fuller models (Battese et al., 1988 &lt;<a href="https://doi.org/10.1080%2F01621459.1988.10478561" target="_top">doi:10.1080/01621459.1988.10478561</a>&gt;). Features include empirical best linear unbiased prediction (EBLUP), analytical and bootstrap Mean Squared Error (MSE) estimation, automatic handling of unsampled domains, and modern S3 diagnostic methods.

Parses compact numeric values, qualified quantities, and ranges commonly found in Chinese tables and spreadsheets. It handles Chinese magnitude suffixes, currencies, percentages, full-width characters, financial negatives, and configurable missing-value markers while reporting values that cannot be parsed safely.

Exploring, visualizing, and comparing classification decision boundaries. Provides a unified interface for fitting classifiers and rendering 2D decision boundary plots, with support for 2D slice visualization (fixing non-plotted dimensions at reference values) and projection-based visualization for high-dimensional data (including Principal Component Analysis (PCA) and tour projections from the 'tourr' package). Supports native 'R' classifiers, 'tidymodels' workflows, and custom user-supplied models via a flexible adapter system. Includes an interactive 'Shiny' application ('explorapp') for visual exploration, data simulation, drawing, model comparison, probability surfaces, and reproducible exports.

Recreation model that simulates household types, group types, expected visiting duration, and total visits from city zip codes to surrounding destination areas. The methodology is based on the accessibility model described in Bervaes et al. (1996) "Een model voor het gebruik van de groene ruimte in stadslandschappen (Fase I)" &lt;<a href="https://www.wur.nl/en/library" target="_top">https://www.wur.nl/en/library</a>&gt;, with optimization procedures based on Goldfarb and Idnani (1983) &lt;<a href="https://doi.org/10.1007%2FBF02591962" target="_top">doi:10.1007/BF02591962</a>&gt; and Vanderbei et al. (1986) &lt;<a href="https://doi.org/10.1007%2FBF01840454" target="_top">doi:10.1007/BF01840454</a>&gt;.

Downloads, reads and organizes data on federal fund transfers to Brazilian municipalities. Data are retrieved from the TransfereGov "Fundo a Fundo" API, from the Portal da Transparência open data service and from the Tesouro Transparente CKAN catalogue. The package offers one reader per API endpoint, helpers that map SIAFI municipality codes to IBGE codes, and routines that fetch the monthly transfer archives and the tax waiver records published as open data.

Estimates latent Gaussian joint distributions for normal continuous, binary, and ordinal variables using marginal summaries from independent studies of a common population. Fits a pairwise Gaussian working criterion using exact summary moments, with study-level sandwich uncertainty. Supports prespecified independent groups, joint event probabilities, and synthetic patient generation. Identification requires repeated joint reporting of variable pairs; heterogeneous populations, rare categories, and small study collections require caution.

Format numbers, percentages and currencies, and label 'ggplot2' axes, using country- or locale-specific conventions such as the thousands separator, decimal mark, currency symbol placement, and sign placement. Locale data is sourced from the Unicode Common Locale Data Repository (CLDR, &lt;<a href="https://cldr.unicode.org" target="_top">https://cldr.unicode.org</a>&gt;) via the 'i18n' package, covering several hundred locales in addition to dedicated helpers for Germany, Switzerland and the United States.

Provides semiparametric smooth-coefficient stochastic frontier analysis following Sun and Kumbhakar (2013) &lt;<a href="https://doi.org/10.1016%2Fj.econlet.2013.05.001" target="_top">doi:10.1016/j.econlet.2013.05.001</a>&gt; where the coefficients of the parametric part vary smoothly with a set of nonparametric variables. Inefficiency term is allowed to depend on a set of determinants through heteroskedasticity. Smooth coefficients are estimated using nonparametric regression and the remaining frontier parameters are estimated by maximum likelihood. Technical efficiency and inefficiency are computed using the Battese and Coelli (1988) &lt;<a href="https://doi.org/10.1016%2F0304-4076%2888%2990053-X" target="_top">doi:10.1016/0304-4076(88)90053-X</a>&gt; and Jondrow et al. (1982) &lt;<a href="https://doi.org/10.1016%2F0304-4076%2882%2990004-5" target="_top">doi:10.1016/0304-4076(82)90004-5</a>&gt; methods, respectively. Confidence intervals for technical efficiency are computed using the approach of Horrace and Schmidt (1996) &lt;<a href="https://doi.org/10.1007%2FBF00157044" target="_top">doi:10.1007/BF00157044</a>&gt;.

Uses the 'Python' implementation of 'Biogeme' as the numerical backend for specifying and estimating discrete-choice models in R. The default native requirement is 'biogeme==3.3.5'.

An end-to-end implementation of the semi-structural quarterly projection models used in central-bank forecasting and policy analysis systems: model declaration with model-consistent expectations, a generalized Schur solver with Blanchard-Kahn diagnostics following Klein (2000) &lt;<a href="https://doi.org/10.1016%2FS0165-1889%2899%2900045-7" target="_top">doi:10.1016/S0165-1889(99)00045-7</a>&gt;, Kalman filtering and smoothing for latent states such as the output gap and the neutral rate, historical shock decompositions, conditional forecasts that distinguish announced from unanticipated policy paths, an auditable judgment ledger, forecast rounds with revision decompositions, Bayesian estimation with identification diagnostics following Iskrev (2010) &lt;<a href="https://doi.org/10.1016%2Fj.jmoneco.2009.12.007" target="_top">doi:10.1016/j.jmoneco.2009.12.007</a>&gt;, and reporting. The canonical small open economy model of Berg, Karam and Laxton (2006) &lt;<a href="https://doi.org/10.5089%2F9781451863413.001" target="_top">doi:10.5089/9781451863413.001</a>&gt; ships as a calibrated template, with extension blocks for disaggregated food inflation and managed exchange rates.