This package performs variable selection for many types of L1-regularised regressions using the revisited knockoffs procedure. This procedure uses a matrix of knockoffs of the covariates independent from the response variable Y. The idea is to determine if a covariate belongs to the model depending on whether it enters the model before or after its knockoff. The procedure suits for a wide range of regressions with various types of response variables. Regression models available are exported from the R packages glmnet and ordinalNet'. Based on the paper linked to via the URL below: Gegout A., Gueudin A., Karmann C. (2019) <arXiv:1907.03153>.

This package provides methods to extract information on pathways, genes and various single-nucleotid polymorphisms (SNPs) from online databases. It provides functions for data preparation and evaluation of genetic influence on a binary outcome using the logistic kernel machine test (LKMT). Three different kernel functions are offered to analyze genotype information in this variance component test: A linear kernel, a size-adjusted kernel and a network-based kernel).

Clustering typically assigns data points into discrete groups, but the clusters can sometimes be indistinct. Cluster sharpening adjusts an existing clustering to create contrast between groups. This package provides a general interface for cluster sharpening along with several implementations based on different excision criteria.

Various tools and data sets that support the study of kanji, including their morphology, decomposition and concepts of distance and similarity between them.

Rcpp implementation of the multivariate Kim filter, which combines the Kalman and Hamilton filters for state probability inference. The filter is designed for state space models and can handle missing values and exogenous data in the observation and state equations. Kim, Chang-Jin and Charles R. Nelson (1999) "State-Space Models with Regime Switching: Classical and Gibbs-Sampling Approaches with Applications" <doi:10.7551/mitpress/6444.001.0001><http://econ.korea.ac.kr/~cjkim/>.

Producing kernel estimates of the unconditional and conditional hazard function for right-censored data including methods of bandwidth selection.

This package provides a function called COTUCKER3() (Co-Inertia Analysis + Tucker3 method) which performs a Co-Tucker3 analysis of two sequences of matrices, as well as other functions called PCA() (Principal Component Analysis) and BGA() (Between-Groups Analysis), which perform analysis of one matrix, COIA() (Co-Inertia Analysis), which performs analysis of two matrices, PTA() (Partial Triadic Analysis), STATIS(), STATISDUAL() and TUCKER3(), which perform analysis of a sequence of matrices, and BGCOIA() (Between-Groups Co-Inertia Analysis), STATICO() (STATIS method + Co-Inertia Analysis), COSTATIS() (Co-Inertia Analysis + STATIS method), which also perform analysis of two sequences of matrices.

Statistical methods that quantify the conditions necessary to alter inferences, also known as sensitivity analysis, are becoming increasingly important to a variety of quantitative sciences. A series of recent works, including Frank (2000) <doi:10.1177/0049124100029002001> and Frank et al. (2013) <doi:10.3102/0162373713493129> extend previous sensitivity analyses by considering the characteristics of omitted variables or unobserved cases that would change an inference if such variables or cases were observed. These analyses generate statements such as "an omitted variable would have to be correlated at xx with the predictor of interest (e.g., the treatment) and outcome to invalidate an inference of a treatment effect". Or "one would have to replace pp percent of the observed data with nor which the treatment had no effect to invalidate the inference". We implement these recent developments of sensitivity analysis and provide modules to calculate these two robustness indices and generate such statements in R. In particular, the functions konfound(), pkonfound() and mkonfound() allow users to calculate the robustness of inferences for a user's own model, a single published study and multiple studies respectively.

This package provides tools for working with the Korea Standard Industrial Classification (KSIC). Includes datasets for the 9th, 10th, and 11th revisions. Functions include searching codes and names by keyword, converting codes across revisions, validating KSIC codes, and navigating the classification hierarchy (e.g., identifying parent or child categories). Intended for use in statistical analysis, data processing, and research involving South Koreaâ s industrial classification system.

Software for k-means clustering of partially observed data from Chi, Chi, and Baraniuk (2016) <doi:10.1080/00031305.2015.1086685>.

Kendall random walks are a continuous-space Markov chains generated by the Kendall generalized convolution. This package provides tools for simulating these random walks and studying distributions related to them. For more information about Kendall random walks see Jasiulis-GoÅ dyn (2014) <arXiv:1412.0220>.

This package provides a fast and computationally efficient algorithm designed to enable researchers to efficiently and quickly extract semantically-related keywords using a fitted embedding model. For more details about the methods applied, see Chester (2025). <doi:10.17605/OSF.IO/5B7RQ>.

This package provides functions for analysing eye tracking data, including event detection, visualizations and area of interest (AOI) based analyses. The package includes implementations of the IV-T, I-DT, adaptive velocity threshold, and Identification by two means clustering (I2MC) algorithms. See separate documentation for each function. The principles underlying I-VT and I-DT algorithms are described in Salvucci & Goldberg (2000) <doi:10.1145/355017.355028>. Two-means clustering is described in Hessels et al. (2017), <doi: 10.3758/s13428-016-0822-1>. The adaptive velocity threshold algorithm is described in Nyström & Holmqvist (2010),<doi:10.3758/BRM.42.1.188>. A documentation of the kollaR can be found in Kleberg et al (2026) <doi:10.3758/s13428-025-02903-z>. Cite this paper when using kollaR See a demonstration in the URL.

Wait for a single key press at the R prompt. This works in terminals, but does not currently work in the Windows GUI', the OS X GUI ('R.app'), in Emacs ESS', in an Emacs shell buffer or in R Studio'. In these cases keypress stops with an error message.

The kernelized version of principal component analysis (KPCA) has proven to be a valid nonlinear alternative for tackling the nonlinearity of biological sample spaces. However, it poses new challenges in terms of the interpretability of the original variables. kpcaIG aims to provide a tool to select the most relevant variables based on the kernel PCA representation of the data as in Briscik et al. (2023) <doi:10.1186/s12859-023-05404-y>. It also includes functions for 2D and 3D visualization of the original variables (as arrows) into the kernel principal components axes, highlighting the contribution of the most important ones.

This package provides the ability to create dynamic citations in which the bibliographic information is pulled from the web rather than having to be entered into a local database such as bibtex ahead of time. The package is primarily aimed at authoring in the R markdown format, and can provide outputs for web-based authoring such as linked text for inline citations. Cite using a DOI', URL, or bibtex file key. See the package URL for details.

Identification of putative causal variants in genome-wide association studies with trio and duo families. The package calculates the W feature statistics from KnockoffTrio and p-values from the family-based association test (FBAT) using trio and/or duo data. Compared to previous versions, a significant improvement has been made in Version 1.1.0 to allow the package to be applied not only to trio families but also to duo families. The package implements the methods in the paper: "Yang, Y., Wang, C., Liu, L., Buxbaum, J., He, Z., & Ionita-Laza, I. (2022). KnockoffTrio: A knockoff framework for the identification of putative causal variants in genome-wide association studies with trio design. The American Journal of Human Genetics, 109(10), 1761-1776.".

In self-reported or anonymised data the user often encounters heaped data, i.e. data which are rounded (to a possibly different degree of coarseness). While this is mostly a minor problem in parametric density estimation the bias can be very large for non-parametric methods such as kernel density estimation. This package implements a partly Bayesian algorithm treating the true unknown values as additional parameters and estimates the rounding parameters to give a corrected kernel density estimate. It supports various standard bandwidth selection methods. Varying rounding probabilities (depending on the true value) and asymmetric rounding is estimable as well: Gross, M. and Rendtel, U. (2016) (<doi:10.1093/jssam/smw011>). Additionally, bivariate non-parametric density estimation for rounded data, Gross, M. et al. (2016) (<doi:10.1111/rssa.12179>), as well as data aggregated on areas is supported.

Cubic spline fitting along with knot selection, includes support for additional variables.

This package provides a collection of personal helper functions to avoid redundancy in the spirit of the "Don't repeat yourself" principle of software development (<https://en.wikipedia.org/wiki/Don%27t_repeat_yourself>).

Implementation of Discrete Symmetric Optimal Kernel for estimating count data distributions, as described by T. Senga Kiessé and G. Durrieu (2024) <doi:10.1016/j.spl.2024.110078>.The nonparametric estimator using the discrete symmetric optimal kernel was illustrated on simulated data sets and a real-word data set included in the package, in comparison with two other discrete symmetric kernels.

This package provides a high-performance R interface to the kuzu graph database. It uses the reticulate package to wrap the official Python client ('kuzu', pandas', and networkx'), allowing users to interact with kuzu seamlessly from within R'. Key features include managing database connections, executing Cypher queries, and efficiently loading data from R data frames. It also provides seamless integration with the R ecosystem by converting query results directly into popular R data structures, including tibble', igraph', tidygraph', and g6R objects, making kuzu's powerful graph computation capabilities readily available for data analysis and visualization workflows in R'. The kuzu documentation can be found at <https://kuzudb.github.io/docs/>.

This package provides a set of tools to analyze texts. Includes, amongst others, functions for automatic language detection, hyphenation, several indices of lexical diversity (e.g., type token ratio, HD-D/vocd-D, MTLD) and readability (e.g., Flesch, SMOG, LIX, Dale-Chall). Basic import functions for language corpora are also provided, to enable frequency analyses (supports Celex and Leipzig Corpora Collection file formats) and measures like tf-idf. Note: For full functionality a local installation of TreeTagger is recommended. It is also recommended to not load this package directly, but by loading one of the available language support packages from the l10n repository <https://undocumeantit.github.io/repos/l10n/>. koRpus also includes a plugin for the R GUI and IDE RKWard, providing graphical dialogs for its basic features. The respective R package rkward cannot be installed directly from a repository, as it is a part of RKWard. To make full use of this feature, please install RKWard from <https://rkward.kde.org> (plugins are detected automatically). Due to some restrictions on CRAN, the full package sources are only available from the project homepage. To ask for help, report bugs, request features, or discuss the development of the package, please subscribe to the koRpus-dev mailing list (<https://korpusml.reaktanz.de>).

This package provides a higher-level interface to the torch package for defining, training, and fine-tuning neural networks, including its depth, powered by code generation. This package supports few to several architectures, including feedforward (multi-layer perceptron) and recurrent neural networks (Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU)), while also reduces boilerplate torch code while enabling seamless integration with torch'. The model methods to train neural networks from this package also bridges to titanic ML frameworks in R, namely tidymodels ecosystem, which enables the parsnip model specifications, workflows, recipes, and tuning tools.