Build decision trees and random forests for classification and regression. The implementation strikes a balance between minimizing computing efforts and maximizing the expected predictive accuracy, thus scales well to large data sets. Multi-threading is available through OpenMP <https://gcc.gnu.org/wiki/openmp>.

This package implements Bayesian model averaging for dynamic panels with weakly exogenous regressors as described in the paper by Moral-Benito (2013, <doi:10.1080/07350015.2013.818003>). The package provides functions to estimate dynamic panel data models and analyze the results of the estimation.

Estimates the causal decompositions of group disparities developed by Yu and Elwert (2025) <doi:10.1214/24-AOAS1990>. For the nuisance functions of the estimators, we provide both parametric and nonparametric options, as well as manual options in case the default models are not satisfying.

Generates all necessary C functions allowing the user to work with the compiled-code interface of ode() and bvptwp(). The implementation supports "forcings" and "events". Also provides functions to symbolically compute Jacobians, sensitivity equations and adjoint sensitivities being the basis for sensitivity analysis.

This function conducts the Cochran-Armitage trend test to a 2 by k contingency table. It will report the test statistic (Z) and p-value.A linear trend in the frequencies will be calculated, because the weights (0,1,2) will be used by default.

Estimates latent variables of public opinion cross-nationally and over time from sparse and incomparable survey data. DCPO uses a population-level graded response model with country-specific item bias terms. Sampling is conducted with Stan'. References: Solt (2020) <doi:10.31235/osf.io/d5n9p>.

This package provides the Empirical Bayesian Elastic Net for handling multicollinearity in generalized linear regression models. As a special case of the EBglmnet package (also available on CRAN), this package encourages a grouping effects to select relevant variables and estimate the corresponding non-zero effects.

This package provides a methodology simple and trustworthy for the analysis of extreme values and multiple threshold tests for a generalized Pareto distribution, together with an automatic threshold selection algorithm. See del Castillo, J, Daoudi, J and Lockhart, R (2014) <doi:10.1111/sjos.12037>.

Given exposure and survival time series as well as parameter values, GUTS allows for the fast calculation of the survival probabilities as well as the logarithm of the corresponding likelihood (see Albert, C., Vogel, S. and Ashauer, R. (2016) <doi:10.1371/journal.pcbi.1004978>).

Generates experiments - simulating structured or experimental data as: completely randomized design, randomized block design, latin square design, factorial and split-plot experiments (Ferreira, 2008, ISBN:8587692526; Naes et al., 2007 <doi:10.1002/qre.841>; Rencher et al., 2007, ISBN:9780471754985; Montgomery, 2001, ISBN:0471316490).

This package provides a collection of functions for testing randomness (or mutual independence) in linear and circular data as proposed in Gehlot and Laha (2025a) <doi:10.48550/arXiv.2506.21157> and Gehlot and Laha (2025b) <doi:10.48550/arXiv.2506.23522>, respectively.

Set of R functions to be coupled with the xeus-r jupyter kernel in order to drive execution of code in notebook input cells, how R objects are to be displayed in output cells, and handle two way communication with the front end through comms.

Run multiple Large Language Model predictions against a table. The predictions run row-wise over a specified column. It works using a one-shot prompt, along with the current row's content. The prompt that is used will depend of the type of analysis needed.

This package provides functions to construct a normative latent cognitive structure (N-LCS) from cognitive test data standardized to healthy controls, and to compute cognitive deviation magnitude (CDM) and cognitive deviation angle (CDA). Methods are described in Chen (2026) <doi:10.1080/23279095.2026.2691088>.

Calibrate p-values under a robust perspective using the methods developed by Sellke, Bayarri, and Berger (2001) <doi:10.1198/000313001300339950> and obtain measures of the evidence provided by the data in favor of point null hypotheses which are safer and more straightforward to interpret.

This package implements the Bayesian quantile regression model for binary longitudinal data (QBLD) developed in Rahman and Vossmeyer (2019) <DOI:10.1108/S0731-90532019000040B009>. The model handles both fixed and random effects and implements both a blocked and an unblocked Gibbs sampler for posterior inference.

This package provides tools for comparing groups using sparse principal component analysis (SPCA). Includes functions for identifying and visualizing influential variables, assessing variable selection stability through bootstrapping, and exploring associations with clinical traits. SPCA is implemented using the method described in <doi:10.1198/106186006X113430>.

Computes the exact joint sparsity and group sparsity projection using the Semi-symmetric Sparse Projection algorithm of J. Shen and S. Damadi, "Sparse projection onto semi-symmetric sets with applications to sparse optimization", Journal of Global Optimization (2026), <doi:10.1007/s10898-026-01592-y>.

This package provides methods for computing joint tests, controlling the Familywise Error Rate (FWER) and getting lower bounds on the number of false hypotheses in a set. The methods implemented here are described in Mogensen and Markussen (2021) <doi:10.48550/arXiv.2108.04731>.

Framework provides functions to parse Training Center XML (TCX) files and extract key activity metrics such as total distance, total time, calories burned, maximum altitude, and power values (watts). This package is useful for analyzing workout and training data from devices that export TCX format.

An interface between R and the Valhalla API. Valhalla is a routing service based on OpenStreetMap data. See <https://valhalla.github.io/valhalla/> for more information. This package enables the computation of routes, trips, isochrones and travel distances matrices (travel time and kilometer distance).

This package provides functions for a classification method based on receiver operating characteristics (ROC). Briefly, features are selected according to their ranked AUC value in the training set. The selected features are merged by the mean value to form a meta-gene. The samples are ranked by their meta-gene value and the meta-gene threshold that has the highest accuracy in splitting the training samples is determined. A new sample is classified by its meta-gene value relative to the threshold. In the first place, the package is aimed at two class problems in gene expression data, but might also apply to other problems.

Scaffolds packages backed by Zig and generates a safe native boundary for plain Zig functions. The generated integration converts objects at the language boundary, registers native routines, propagates errors, and compiles across platforms without requiring C++. Each step of the workflow is a function: locating a supported Zig compiler, scaffolding the package, scanning the Zig source for exported functions, rendering and writing the bindings, checking the package before compilation, and building it with the compiler errors reported by file and line. The scan, the checks and the generated bindings are pure R; a Zig compiler is needed only to build a package created with the scaffold.

Raw vectors in R are useful for storing a single binary object. What if you want to put a vector of them in a data frame? The blob package provides the blob object, a list of raw vectors, suitable for use as a column in data frame.