Stability based methods for model order selection in clustering problems (Valentini, G (2007), <doi:10.1093/bioinformatics/btl600>). Using multiple perturbations of the data the stability of clustering solutions is assessed. Different perturbations may be used: resampling techniques, random projections and noise injection. Stability measures for the estimate of clustering solutions and statistical tests to assess their significance are provided.

This package provides multigroup Kitagawa-Blinder-Oaxaca ('mKBO') decompositions, that allow for more than two groups. Each group is compared to the sample average. For more details see Thaning and Nieuwenhuis (2025) <doi:10.31235/osf.io/6twvj_v1>.

Computes a finite-sample tail bound for the log-likelihood ratio test (LRT) statistic under multinomial sampling. The resulting bound is used to compute finite-sample conservative p-values and critical values when the standard chi-squared asymptotics can be unreliable. The package also supports multiple independent multinomial trials.

Renders rich inline text (bold, italic, code, links, images) in grid graphics and ggplot2', from markdown or flextable chunks. Provides grobs, theme elements, and geometry layers for styled text rendering. Only works with graphics devices that support systemfonts', such as those provided by ragg', svglite', or ggiraph'. The cairo_pdf device is also supported when fonts are installed at the system level.

Software to support the introductory *MOSAIC Calculus* textbook <https://www.mosaic-web.org/MOSAIC-Calculus/>), one of many data- and modeling-oriented educational resources developed by Project MOSAIC (<https://www.mosaic-web.org/>). Provides symbolic and numerical differentiation and integration, as well as support for applied linear algebra (for data science), and differential equations/dynamics. Includes grammar-of-graphics-based functions for drawing vector fields, trajectories, etc. The software is suitable for general use, but intended mainly for teaching calculus.

Aggregates matrix population models (MPMs) in both the lambda (stable growth rate) and R0 (net reproductive rate) frameworks, including standard and elasticity-consistent aggregators. Standard aggregation in the lambda framework maintains consistent lambda and stable stage distribution, while standard aggregation in the R0 framework maintains consistent R0 and cohort stable stage distribution. Elasticity-consistent aggregators maintain these same consistencies with respect to the chosen framework and additionally preserve consistent reproductive values in the lambda framework and cohort reproductive values in the R0 framework. Aggregation can take the form of general-to-general MPM (mpm_aggregate) or Leslie-to-Leslie MPM (leslie_aggregate).

We develop Multi-source Graph Synthesis (MUGS), an algorithm designed to create embeddings for pediatric Electronic Health Record (EHR) codes by leveraging graphical information from three distinct sources: (1) pediatric EHR data, (2) EHR data from the general patient population, and (3) existing hierarchical medical ontology knowledge shared across different patient populations. See Li et al. (2024) <doi:10.1038/s41746-024-01320-4> for details.

This package implements the split-fit-evaluate-assess workflow from Hastie, Tibshirani, and Friedman (2009, ISBN:978-0-387-84857-0) "The Elements of Statistical Learning", Chapter 7. Provides three-way data splitting with automatic stratification, mandatory seeds for reproducibility, automatic data type handling, and 10 algorithms out of the box. Uses Rust backend for cross-language deterministic splitting. Designed for tabular supervised learning with minimal ceremony. Polyglot parity with the Python mlw package on PyPI'.

This package performs multiple imputation of missing data using an ensemble super learner built with the tidymodels framework. For each incomplete column, a stacked ensemble of candidate learners is trained on a bootstrap sample of the observed data and used to generate imputations via predictive mean matching (continuous), probability draws (binary), or cumulative probability draws (categorical). Supports parallelism across imputed datasets via the future framework.

Fit data from a continuous population with a smooth density on finite interval by an approximate Bernstein polynomial model which is a mixture of certain beta distributions and find maximum approximate Bernstein likelihood estimator of the unknown coefficients. Consequently, maximum likelihood estimates of the unknown density, distribution functions, and more can be obtained. If the support of the density is not the unit interval then transformation can be applied. This is an implementation of the methods proposed by the author of this package published in the Journal of Nonparametric Statistics: Guan (2016) <doi:10.1080/10485252.2016.1163349> and Guan (2017) <doi:10.1080/10485252.2017.1374384>. For data with covariates, under some semiparametric regression models such as Cox proportional hazards model and the accelerated failure time model, the baseline survival function can be estimated smoothly based on general interval censored data.

Counting process structure is fundamental to model time varying covariates. This package restructures dataframes in the counting process format for one or more variables. F. W. Dekker, et al. (2008) <doi:10.1038/ki.2008.328>.

Fast approximate methods for mixed logistic regression in genome-wide analysis studies (GWAS). Two computationnally efficient methods are proposed for obtaining effect size estimates (beta) in Mixed Logistic Regression in GWAS: the Approximate Maximum Likelihood Estimate (AMLE), and the Offset method. The wald test obtained with AMLE is identical to the score test. Data can be genotype matrices in plink format, or dosage (VCF files). The methods are described in details in Milet et al (2020) <doi:10.1101/2020.01.17.910109>.

Fits Semiparametric Promotion Time Cure Models, taking into account (using a corrected score approach or the SIMEX algorithm) or not the measurement error in the covariates, using a backfitting approach to maximize the likelihood.

This package provides functions for detecting multicollinearity. This test gives statistical support to two of the most famous methods for detecting multicollinearity in applied work: Kleinâ s rule and Variance Inflation Factor (VIF). See the URL for the papers associated with this package, as for instance, Morales-Oñate and Morales-Oñate (2015) <doi:10.33333/rp.vol51n2.05>.

This package provides a tool for optimizing scales of effect when modeling ecological processes in space. Specifically, the scale parameter of a distance-weighted kernel distribution is identified for all environmental layers included in the model. Includes functions to assist in model selection, model evaluation, efficient transformation of raster surfaces using fast Fourier transformation, and projecting models. For more details see Peterman (2026) <doi:10.1007/s10980-025-02267-x>.

Discrete event simulation using both R and C++ (Karlsson et al 2016; <doi:10.1109/eScience.2016.7870915>). The C++ code is adapted from the SSIM library <https://www.inf.usi.ch/carzaniga/ssim/>, allowing for event-oriented simulation. The code includes a SummaryReport class for reporting events and costs by age and other covariates. The C++ code is available as a static library for linking to other packages. A priority queue implementation is given in C++ together with an S3 closure and a reference class implementation. Finally, some tools are provided for cost-effectiveness analysis.

Performing multiple-class cluster correspondence analysis(MCCCA). The main functions are create.MCCCAdata() to create a list to be applied to MCCCA, MCCCA() to apply MCCCA, and plot.mccca() for visualizing MCCCA result. Methods used in the package refers to Mariko Takagishi and Michel van de Velden (2022)<doi:10.1080/10618600.2022.2035737>.

Test for monotonicity in financial variables sorted by portfolios. It is conventional practice in empirical research to form portfolios of assets ranked by a certain sort variable. A t-test is then used to consider the mean return spread between the portfolios with the highest and lowest values of the sort variable. Yet comparing only the average returns on the top and bottom portfolios does not provide a sufficient way to test for a monotonic relation between expected returns and the sort variable. This package provides nonparametric tests for the full set of monotonic patterns by Patton, A. and Timmermann, A. (2010) <doi:10.1016/j.jfineco.2010.06.006> and compares the proposed results with extant alternatives such as t-tests, Bonferroni bounds, and multivariate inequality tests through empirical applications and simulations.

This package implements parametric modal regression for continuous positive distributions of the exponential family under right censoring. Provides functions to link the conditional mode to a linear predictor using reparameterizations for Gamma, Beta, Weibull, and Inverse Gaussian families. Includes maximum likelihood estimation via numerical optimization, asymptotic inference based on the observed Fisher information matrix, and model diagnostics using randomized quantile residuals.

This package implements the algorithm of Remez (1962) for polynomial minimax approximation and of Cody et al. (1968) <doi:10.1007/BF02162506> for rational minimax approximation.

Various reliability analysis methods for rare event inference (computing failure probability and quantile from model/function outputs).

Gene Expression datasets for the MM2S package. Contains normalized expression data for Human Medulloblastoma ('GSE37418') as well as Mouse Medulloblastoma models ('GSE36594'). Deena Gendoo et al. (2015) <doi:10.1016/j.ygeno.2015.05.002>.

Estimation, inference and diagnostics for Univariate Autoregressive Markov Switching Models for Linear and Generalized Models. Distributions for the series include gaussian, Poisson, binomial and gamma cases. The EM algorithm is used for estimation (see Perlin (2012) <doi:10.2139/ssrn.1714016>).

This package provides tools to help convert credit risk data at two timepoints into traditional credit state migration (aka, "transition") matrices. At a higher level, migrate is intended to help an analyst understand how risk moved in their credit portfolio over a time interval. References to this methodology include: 1. Schuermann, T. (2008) <doi:10.1002/9780470061596.risk0409>. 2. Perederiy, V. (2017) <doi:10.48550/arXiv.1708.00062>.