Fit various smoothing spline models. Includes an ssr() function for smoothing spline regression, an nnr() function for nonparametric nonlinear regression, an snr() function for semiparametric nonlinear regression, an slm() function for semiparametric linear mixed-effects models, and an snm() function for semiparametric nonlinear mixed-effects models. See Wang (2011) <doi:10.1201/b10954> for an overview.

Create, upload and run Acumos R models. Acumos (<https://www.acumos.org>) is a platform and open source framework intended to make it easy to build, share, and deploy AI apps. Acumos is part of the LF AI Foundation', an umbrella organization within The Linux Foundation'. With this package, user can create a component, and push it to an Acumos platform.

This package implements Bayesian inference to detect signal from blinded clinical trial when total number of adverse events of special concerns and total risk exposures from all patients are available in the study. For more details see the article by Mukhopadhyay et. al. (2018) titled Bayesian Detection of Potential Risk Using Inference on Blinded Safety Data', in Pharmaceutical Statistics (to appear).

This package provides functions to compute the joint probability mass function (pmf), cumulative distribution function (cdf), and survival function (sf) of the Basu-Dhar bivariate geometric distribution. Additional functionalities include the calculation of the correlation coefficient, covariance, and cross-factorial moments, as well as the generation of random variates. The package also implements parameter estimation based on the method of moments.

This package provides classes (S4) of circular-linear, symmetric copulas with corresponding methods, extending the copula package. These copulas are especially useful for modeling correlation in discrete-time movement data. Methods for density, (conditional) distribution, random number generation, bivariate dependence measures and fitting parameters using maximum likelihood and other approaches. The package also contains methods for visualizing movement data and copulas.

This package provides a Shiny'-based toolkit for item/test analysis. It is designed for multiple-choice, true-false, and open-ended questions. The toolkit is usable with datasets in 1-0 or other formats. Key analyses include difficulty, discrimination, response-option analysis, and reports. The classical test theory methods used are described in Ebel and Frisbie (1991, ISBN:978-0132892314).

This package provides the analysis of variance table including the expected mean squares (EMS) for various types of experimental design. When some variables are random effects or we use special experimental design such as nested design, repeated-measures design, or split-plot design, it is not easy to find the appropriate test, especially denominator for F-statistic which depends on EMS.

Classical (bottom-up and top-down), optimal combination and heuristic point (Di Fonzo and Girolimetto, 2023 <doi:10.1016/j.ijforecast.2021.08.004>) and probabilistic (Girolimetto et al. 2024 <doi:10.1016/j.ijforecast.2023.10.003>) forecast reconciliation procedures for linearly constrained time series (e.g., hierarchical or grouped time series) in cross-sectional, temporal, or cross-temporal frameworks.

Triangular and trapezoidal fuzzy numbers are used to study fuzzy logic, fuzzy reasoning and approximating, fuzzy regression models, etc. This package builds the generating function for triangular and trapezoidal fuzzy numbers based on Souliotis et al. (2022)<doi:10.3390/math10183350>. They proposed a method for the construction of fuzzy numbers via a cumulative distribution function based on the possibility theory.

Approximate frequentist inference for generalized linear mixed model analysis with expectation propagation used to circumvent the need for multivariate integration. In this version, the random effects can be any reasonable dimension. However, only probit mixed models with one level of nesting are supported. The methodology is described in Hall, Johnstone, Ormerod, Wand and Yu (Journal of the American Statistical Association, 2020).

This package performs a homogeneity analysis (multiple correspondence analysis) and various extensions. Rank restrictions on the category quantifications can be imposed (nonlinear PCA). The categories are transformed by means of optimal scaling with options for nominal, ordinal, and numerical scale levels (for rank-1 restrictions). Variables can be grouped into sets, in order to emulate regression analysis and canonical correlation analysis.

Miscellaneous convenience functions and wrapper functions to convert frequencies between Hz, semitones, mel and Bark, to create a matrix of dummy columns from a factor, to determine whether x lies in range [a,b], and to add a bracketed line to an existing plot. This package also contains an example data set of a stratified sample of 80 talkers of Dutch.

Select set of parametric and non-parametric statistical tests. inferr builds upon the solid set of statistical tests provided in stats package by including additional data types as inputs, expanding and restructuring the test results. The tests included are t tests, variance tests, proportion tests, chi square tests, Levene's test, McNemar Test, Cochran's Q test and Runs test.

Item response theory (IRT) parameter estimation using marginal maximum likelihood and expectation-maximization algorithm (Bock \& Aitkin, 1981 <doi:10.1007/BF02293801>). Within parameter estimation algorithm, several methods for latent distribution estimation are available. Reflecting some features of the true latent distribution, these latent distribution estimation methods can possibly enhance the estimation accuracy and free the normality assumption on the latent distribution.

Modular implementation of Multiobjective Evolutionary Algorithms based on Decomposition (MOEA/D) [Zhang and Li (2007), <DOI:10.1109/TEVC.2007.892759>] for quick assembling and testing of new algorithmic components, as well as easy replication of published MOEA/D proposals. The full framework is documented in a paper published in the Journal of Statistical Software [<doi:10.18637/jss.v092.i06>].

Characterization of distribution data on the surface of a sphere. The primary group of these metrics describe the extent of a distribution, geographic ranges. The calculation of geographic descriptors can be executed using point coordinate data, vector polygons, as well as cells on a discretized sphere. Besides using spherical implementations, the package offers the exploration of partial results for visual diagnostics.

This package provides tools from the domain of graph theory can be used to quantify the complexity and vulnerability to failure of a software package. That is the guiding philosophy of this package. pkgnet provides tools to analyze the dependencies between functions in an R package and between its imported packages. See the pkgnet website for vignettes and other supplementary information.

Bayesian variable selection for linear regression models using hierarchical priors. There is a prior that combines information across responses and one that combines information across covariates, as well as a standard spike and slab prior for comparison. An MCMC samples from the marginal posterior distribution for the 0-1 variables indicating if each covariate belongs to the model for each response.

This package provides the Fortran code of the R package spam with 64-bit integers. Loading this package together with the R package spam enables the sparse matrix class spam to handle huge sparse matrices with more than 2^31-1 non-zero elements. Documentation is provided in Gerber, Moesinger and Furrer (2017) <doi:10.1016/j.cageo.2016.11.015>.

Tipping point analysis for clinical trials that employ Bayesian dynamic borrowing via robust meta-analytic predictive (MAP) priors. Further functions facilitate expert elicitation of a primary weight of the informative component of the robust MAP prior and computation of operating characteristics. Intended use is the planning, analysis and interpretation of extrapolation studies in pediatric drug development, but applicability is generally wider.

This package provides a framework to estimate high dimensional generalized structural equation models using two stage quasi-likelihood expectation-maximization. The structural model supports binomial (logit and probit), Poisson, negative binomial, and gamma distributions for the outcome variable. Hattab (2026) "A Two Stage Quasi-Likelihood Estimation Method for High Dimensional Generalized Structural Equation Models" <doi:10.48550/arXiv.2608.16017>.

Repo is a tool built on top of Git. Repo helps manage many Git repositories, does the uploads to revision control systems, and automates parts of the development workflow. Repo is not meant to replace Git, only to make it easier to work with Git. The repo command is an executable Python script that you can put anywhere in your path.

Collection of methods for rating matrix completion, which is a statistical framework for recommender systems. Another relevant application is the imputation of rating-scale survey data in the social and behavioral sciences. Note that matrix completion and imputation are synonymous terms used in different streams of the literature. The main functionality implements robust matrix completion for discrete rating-scale data with a low-rank constraint on a latent continuous matrix (Archimbaud, Alfons, and Wilms (2025) <doi:10.48550/arXiv.2412.20802>). In addition, the package provides wrapper functions for softImpute (Mazumder, Hastie, and Tibshirani, 2010, <https://www.jmlr.org/papers/v11/mazumder10a.html>; Hastie, Mazumder, Lee, Zadeh, 2015, <https://www.jmlr.org/papers/v16/hastie15a.html>) for easy tuning of the regularization parameter, as well as benchmark methods such as median imputation and mode imputation.

The functions in this package compute robust estimators by minimizing a kernel-based distance known as MMD (Maximum Mean Discrepancy) between the sample and a statistical model. Recent works proved that these estimators enjoy a universal consistency property, and are extremely robust to outliers. Various optimization algorithms are implemented: stochastic gradient is available for most models, but the package also allows gradient descent in a few models for which an exact formula is available for the gradient. In terms of distribution fit, a large number of continuous and discrete distributions are available: Gaussian, exponential, uniform, gamma, Poisson, geometric, etc. In terms of regression, the models available are: linear, logistic, gamma, beta and Poisson. Alquier, P. and Gerber, M. (2024) <doi:10.1093/biomet/asad031> Cherief-Abdellatif, B.-E. and Alquier, P. (2022) <doi:10.3150/21-BEJ1338>.