Four methods for mediation analysis with missing data: Listwise deletion, Pairwise deletion, Multiple imputation, and Two Stage Maximum Likelihood algorithm. For MI and TS-ML, auxiliary variables can be included. Bootstrap confidence intervals for mediation effects are obtained. The robust method is also implemented for TS-ML. Since version 1.4, bmem adds the capability to conduct power analysis for mediation models. Details about the methods used can be found in these articles. Zhang and Wang (2003) <doi:10.1007/s11336-012-9301-5>. Zhang (2014) <doi:10.3758/s13428-013-0424-0>.

Fit latent space network cluster models using an expectation-maximization algorithm. Enables flexible modeling of unweighted or weighted network data (with or without noise edges), supporting both directed and undirected networks (with or without degree and strength heterogeneity). Designed to handle large networks efficiently, it allows users to explore network structure through latent space representations, identify clusters (i.e., community detection) within network data, and simulate networks with varying clustering, connectivity patterns, and noise edges. Methodology for the implementation is described in Arakkal and Sewell (2025) <doi:10.1016/j.csda.2025.108228>.

When building complex models, it is often difficult to explain why the model should be trusted. While global measures such as accuracy are useful, they cannot be used for explaining why a model made a specific prediction. lime (a port of the lime Python package) is a method for explaining the outcome of black box models by fitting a local model around the point in question an perturbations of this point. The approach is described in more detail in the article by Ribeiro et al. (2016) <doi:10.48550/arXiv.1602.04938>.

This package implements the Panel Smooth Transition Regression (PSTR) framework for nonlinear panel data modelling. The modelling procedure consists of three stages: Specification, Estimation and Evaluation. The package provides tools for model specification testing, to do PSTR model estimation, and to do model evaluation. The implemented tests allow for cluster dependence and are heteroskedasticity-consistent. The wild bootstrap and wild cluster bootstrap tests are also implemented. Parallel computation (as an option) is implemented in some functions, especially the bootstrap tests. The package supports parallel computation, which is useful for large-scale bootstrap procedures.

We provide functionality to implement penalized PCA with an option to smooth the objective function using Nesterov smoothing. Two functions are available to compute a user-specified number of eigenvectors. The function unsmoothed_penalized_EV() computes a penalized PCA without smoothing and has three parameters (the input matrix, the Lasso penalty, and the number of desired eigenvectors). The function smoothed_penalized_EV() computes a smoothed penalized PCA using the same parameters and additionally requires the specification of a smoothing parameter. Both functions return a matrix having the desired eigenvectors as columns.

Implement a GAM-based (Generalized Additive Models) spatial surplus production model (spatial SPM), aimed at modeling northern shrimp population in Atlantic Canada but potentially to any stock in any location. The package is opinionated in its implementation of SPMs as it internally makes the choice to use penalized spatial gams with time lags. However, it also aims to provide options for the user to customize their model. The methods are described in Pedersen et al. (2022, <https://www.dfo-mpo.gc.ca/csas-sccs/Publications/ResDocs-DocRech/2022/2022_062-eng.html>).

Monte Carlo sampling algorithms for semiparametric Bayesian regression analysis. These models feature a nonparametric (unknown) transformation of the data paired with widely-used regression models including linear regression, spline regression, quantile regression, and Gaussian processes. The transformation enables broader applicability of these key models, including for real-valued, positive, and compactly-supported data with challenging distributional features. The samplers prioritize computational scalability and, for most cases, Monte Carlo (not MCMC) sampling for greater efficiency. Details of the methods and algorithms are provided in Kowal and Wu (2024) <doi:10.1080/01621459.2024.2395586>.

Using The Free Evocation of Words Technique method with some functions, this package will make a social representation and other analysis. The Free Evocation of Words Technique consists of collecting a number of words evoked by a subject facing exposure to an inducer term. The purpose of this technique is to understand the relationships created between words evoked by the individual and the inducer term. This technique is included in the theory of social representations, therefore, on the information transmitted by an individual, seeks to create a profile that define a social group.

Publicly available RNA-seq data is routinely used for retrospective analysis to elucidate new biology. Novel transcript discovery enabled by large collections of RNA-seq datasets has emerged as one of such analysis. To increase the power of transcript discovery from large collections of RNA-seq datasets, we developed a new R package named Pooling RNA-seq and Assembling Models (PRAM), which builds transcript models in intergenic regions from pooled RNA-seq datasets. This package includes functions for defining intergenic regions, extracting and pooling related RNA-seq alignments, predicting, selected, and evaluating transcript models.

This package provides methods for estimating the area under the concentration versus time curve (AUC) and its standard error in the presence of Below the Limit of Quantification (BLOQ) observations. Two approaches are implemented: direct estimation using censored maximum likelihood, and a two-step approach that first imputes BLOQ values using various methods and then computes the AUC using the imputed data. Technical details are described in Barnett et al. (2020), "Methods for Non-Compartmental Pharmacokinetic Analysis With Observations Below the Limit of Quantification," Statistics in Biopharmaceutical Research. <doi:10.1080/19466315.2019.1701546>.

Fits a constrained regression model for an ordinal response with ordinal predictors and possibly others, Espinosa and Hennig (2019) <DOI:10.1007/s11222-018-9842-2>. The parameter estimates associated with an ordinal predictor are constrained to be monotonic. If a monotonicity direction (isotonic or antitonic) is not specified for an ordinal predictor by the user, then one of the available methods will either establish it or drop the monotonicity assumption. Two monotonicity tests are also available to test the null hypothesis of monotonicity over a set of parameters associated with an ordinal predictor.

Train a Gaussian stochastic process model of an unknown function, possibly observed with error, via maximum likelihood or maximum a posteriori (MAP) estimation, run model diagnostics, and make predictions, following Sacks, J., Welch, W.J., Mitchell, T.J., and Wynn, H.P. (1989) "Design and Analysis of Computer Experiments", Statistical Science, <doi:10.1214/ss/1177012413>. Perform sensitivity analysis and visualize low-order effects, following Schonlau, M. and Welch, W.J. (2006), "Screening the Input Variables to a Computer Model Via Analysis of Variance and Visualization", <doi:10.1007/0-387-28014-6_14>.

This package implements the Heterogeneous Transition Network Analysis (HTNA) method described by López-Pernas et al. (2026) <doi:10.1002/jcal.70285>. The method is an extension of transition network analysis (TNA) where actions or events belong to two or more distinct actor types (e.g. Human and AI), preserving the actor type partition on the resulting network. Provides a thin, focused API on top of the Nestimate estimation engine and the cograph rendering engine, so downstream bootstrap, permutation, reliability, centrality, and plotting functions treat each actor's codes as a distinct node group.

This package provides a streamlined tool for eplet analysis of donor and recipient HLA (human leukocyte antigen) mismatch. Messy, low-resolution HLA typing data is cleaned, and imputed to high-resolution using the NMDP (National Marrow Donor Program) haplotype reference database <https://haplostats.org/haplostats>. High resolution data is analyzed for overall or single antigen eplet mismatch using a reference table (currently supporting HLAMatchMaker <http://www.epitopes.net> versions 2 and 3). Data can enter or exit the workflow at different points depending on the user's aims and initial data quality.

Create interactive d3.js visualizations from R with built-in statistical transforms. Computes confidence intervals, regression fits, LOESS smoothing, moving averages, error bars, and uncertainty visualizations (quantile dot plots and fan charts) in R and renders them as composable chart layers via htmlwidgets'. Supports 36 chart types including boxplots, violin plots, Q-Q diagnostic plots, calendar heatmaps, survival curves, and group comparisons with pairwise significance testing. Also provides a machine-readable chart specification schema with validators so that large language model agents can author and verify charts. Works in RStudio', Shiny', and R Markdown'.

The oblique decision tree (ODT) uses linear combinations of predictors as partitioning variables in a decision tree. Oblique Decision Random Forest (ODRF) is an ensemble of multiple ODTs generated by feature bagging. Oblique Decision Boosting Tree (ODBT) applies feature bagging during the training process of ODT-based boosting trees to ensemble multiple boosting trees. All three methods can be used for classification and regression, and ODT and ODRF serve as supplements to the classical CART of Breiman (1984) <DOI:10.1201/9781315139470> and Random Forest of Breiman (2001) <DOI:10.1023/A:1010933404324> respectively.

Set of tools to fit a semi-parametric regression model suitable for analysis of data sets in which the response variable is continuous, strictly positive, asymmetric and possibly, censored. Under this setup, both the median and the skewness of the response variable distribution are explicitly modeled by using semi-parametric functions, whose non-parametric components may be approximated by natural cubic splines or P-splines. Supported distributions for the model error include log-normal, log-Student-t, log-power-exponential, log-hyperbolic, log-contaminated-normal, log-slash, Birnbaum-Saunders and Birnbaum-Saunders-t distributions.

Elaboration of vehicular emissions inventories, consisting in four stages, pre-processing activity data, preparing emissions factors, estimating the emissions and post-processing of emissions in maps and databases. More details in Ibarra-Espinosa et al (2018) <doi:10.5194/gmd-11-2209-2018>. Before using VEIN you need to know the vehicular composition of your study area, in other words, the combination of of type of vehicles, size and fuel of the fleet. Then, it is recommended to start with the project to download a template to create a structure of directories and scripts.

This package provides a parallel implementation of Weighted Subspace Random Forest. The Weighted Subspace Random Forest algorithm was proposed in the International Journal of Data Warehousing and Mining by Baoxun Xu, Joshua Zhexue Huang, Graham Williams, Qiang Wang, and Yunming Ye (2012) <DOI:10.4018/jdwm.2012040103>. The algorithm can classify very high-dimensional data with random forests built using small subspaces. A novel variable weighting method is used for variable subspace selection in place of the traditional random variable sampling.This new approach is particularly useful in building models from high-dimensional data.

Estimates the standard and weighted Elo (WElo, Angelini et al., 2022 <doi:10.1016/j.ejor.2021.04.011>) rates. The current version provides Elo and WElo rates for tennis, according to different systems of weights (games or sets) and scale factors (constant, proportional to the number of matches, with more weight on Grand Slam matches or matches played on a specific surface). Moreover, the package gives the possibility of estimating the (bootstrap) standard errors for the rates. Finally, the package includes betting functions that automatically select the matches on which place a bet.

Supports a structured approach for exploring PKPD data <https://opensource.nibr.com/xgx/>. It also contains helper functions for enabling the modeler to follow best R practices (by appending the program name, figure name location, and draft status to each plot). In addition, it enables the modeler to follow best graphical practices (by providing a theme that reduces chart ink, and by providing time-scale, log-scale, and reverse-log-transform-scale functions for more readable axes). Finally, it provides some data checking and summarizing functions for rapidly exploring pharmacokinetics and pharmacodynamics (PKPD) datasets.

We provide an implementation for Sum of Ranking Differences (SRD), a novel statistical test introduced by Héberger (2010) <doi:10.1016/j.trac.2009.09.009>. The test allows the comparison of different solutions through a reference by first performing a rank transformation on the input, then calculating and comparing the distances between the solutions and the reference - the latter is measured in the L1 norm. The reference can be an external benchmark (e.g. an established gold standard) or can be aggregated from the data. The calculated distances, called SRD scores, are validated in two ways, see Héberger and Kollár-Hunek (2011) <doi:10.1002/cem.1320>. A randomization test (also called permutation test) compares the SRD scores of the solutions to the SRD scores of randomly generated rankings. The second validation option is cross-validation that checks whether the rankings generated from the solutions come from the same distribution or not. For a detailed analysis about the cross-validation process see Sziklai, Baranyi and Héberger (2021) <doi:10.48550/arXiv.2105.11939>. The package offers a wide array of features related to SRD including the computation of the SRD scores, validation options, input preprocessing and plotting tools.

This package provides an up-to-date copy of the Internet Assigned Numbers Authority (IANA) Time Zone Database. It is updated periodically to reflect changes made by political bodies to time zone boundaries, UTC offsets, and daylight saving time rules. Additionally, this package provides a C++ interface for working with the date library. date provides comprehensive support for working with dates and date-times, which this package exposes to make it easier for other R packages to utilize. Headers are provided for calendar specific calculations, along with a limited interface for time zone manipulations.

This package implements the Bi-objective Regression Tree (BORT) for efficiently learning vector-valued functions. Unlike traditional methods that rely on constructing multiple models or static scalarisation, BORT integrates the exploration of the Pareto front directly into a single tree's growth process. It provides high-efficiency, single-model approaches that can Pareto-dominate entire Pareto-consistent families of trees, supported by a C backend for fast computation. For more details see Paz (2026) <doi:10.1007/978-3-032-28393-1_2> and Paz (2025) <doi:10.1007/978-3-031-78401-9_2>.