Algorithms of nonparametric sequential test and online change-point detection for streams of univariate (sub-)Gaussian, binary, and bounded random variables, introduced in following publications - Shin et al. (2024) <doi:10.48550/arXiv.2203.03532>, Shin et al. (2021) <doi:10.48550/arXiv.2010.08082>.

This package provides tools for analyzing spatial cell-cell interactions based on ligand-receptor pairs, including functions for local, regional, and global analysis using spatial transcriptomics data. Integrates with databases like CellChat <https://github.com/jinworks/CellChat>, CellPhoneDB <https://www.cellphonedb.org/>, Cellinker <https://www.rna-society.org/cellinker/>, ICELLNET <https://github.com/soumelis-lab/ICELLNET>, and ConnectomeDB <https://humanconnectome.org/software/connectomedb/> to identify ligand-receptor pairs, visualize interactions through heatmaps, chord diagrams, and infer interactions on different spatial scales.

This package implements several functions that automates the cleaning and spell-checking of text data. Also converges, finalizes, removes plurals and continuous strings, and puts text data in binary format for semantic network analysis. Uses the SemNetDictionaries package to make the cleaning process more accurate, efficient, and reproducible.

Analysis of species count data in ecology often requires normalization to an identical sample size. Rarefying (random subsampling without replacement), which is a popular method for normalization, has been widely criticized for its poor reproducibility and potential distortion of the community structure. In the context of microbiome count data, researchers explicitly advised against the use of rarefying. An alternative to rarefying is scaling with ranked subsampling (SRS). SRS consists of two steps. In the first step, the total counts for all OTUs (operational taxonomic units) or species in each sample are divided by a scaling factor chosen in such a way that the sum of the scaled counts Cscaled equals Cmin. In the second step, the non-integer Cscaled values are converted into integers by an algorithm that we dub ranked subsampling. The Cscaled value for each OTU or species is split into the integer part Cint (Cint = floor(Cscaled)) and the fractional part Cfrac (Cfrac = Cscaled - Cints). Since the sum of Cint is smaller or equal to Cmin, additional delta C = Cmin - the sum of Cint counts have to be added to the library to reach the total count of Cmin. This is achieved as follows. OTUs are ranked in the descending order of their Cfrac values. Beginning with the OTU of the highest rank, single count per OTU is added to the normalized library until the total number of added counts reaches delta C and the sum of all counts in the normalized library equals Cmin. When the lowest Cfrag involved in picking delta C counts is shared by several OTUs, the OTUs used for adding a single count to the library are selected in the order of their Cint values. This selection minimizes the effect of normalization on the relative frequencies of OTUs. OTUs with identical Cfrag as well as Cint are sampled randomly without replacement. See Beule & Karlovsky (2020) <doi:10.7717/peerj.9593> for details.

Fits Bayesian hierarchical piecewise regression models with multiple logistic-smoothed change-points. Non-linear parameters (change-point locations and transition sharpness) and linear parameters can each be conditioned on covariates and factors via flexible design matrices. A random-intercept structure is supported for any parameter. Spike-and-slab regularization is supported for selecting the number of breakpoints. Posterior inference uses a Metropolis-within-Gibbs sampler implemented in Rust for speed. Methods are based on the smooth transition piecewise regression model of Bacon and Watts (1971) <doi:10.2307/2334389> and variable selection spike-and-slab priors of Kuo and Mallick (1998) <https://www.jstor.org/stable/25053023>.

This package provides tools for simulating spatially dependent predictors (continuous or binary), which are used to generate scalar outcomes in a (generalized) linear model framework. Continuous predictors are generated using traditional multivariate normal distributions or Gauss Markov random fields with several correlation function approaches (e.g., see Rue (2001) <doi:10.1111/1467-9868.00288> and Furrer and Sain (2010) <doi:10.18637/jss.v036.i10>), while binary predictors are generated using a Boolean model (see Cressie and Wikle (2011, ISBN: 978-0-471-69274-4)). Parameter vectors exhibiting spatial clustering can also be easily specified by the user.

This package provides tools for Genotype by Environment Interaction (GEI) analysis, using statistical models and visualizations to assess genotype performance across environments. It helps researchers explore interaction effects, stability, and adaptability in multi-environment trials, identifying the best-performing genotypes in different conditions. Which Win Where!

Various tools for semantic vector spaces, such as correspondence analysis (simple, multiple and discriminant), latent semantic analysis, probabilistic latent semantic analysis, non-negative matrix factorization, latent class analysis, EM clustering, logratio analysis and log-multiplicative (association) analysis. Furthermore, there are specialized distance measures, plotting functions and some helper functions.

Simulate complex data from a given directed acyclic graph and information about each individual node. Root nodes are simply sampled from the specified distribution. Child Nodes are simulated according to one of many implemented regressions, such as logistic regression, linear regression, poisson regression or any other function. Also includes a comprehensive framework for discrete-time simulation, discrete-event simulation, and networks-based simulation which can generate even more complex longitudinal and dependent data. For more details, see Robin Denz, Nina Timmesfeld (2026) <doi:10.18637/jss.v116.i02>.

Combine topic modeling and sentiment analysis to identify individual students gaps, and highlight their strengths and weaknesses across predefined competency domains and professional activities.

This package provides a pipeline for estimating the stellar age, mass, and radius given observational effective temperature, [Fe/H], and astroseismic parameters. The results are obtained adopting a maximum likelihood technique over a grid of pre-computed stellar models, as described in Valle et al. (2014) <doi:10.1051/0004-6361/201322210>.

This package provides a mostly pure-R implementation of the RAKE algorithm (Rose, S., Engel, D., Cramer, N. and Cowley, W. (2010) <doi:10.1002/9780470689646.ch1>), which can be used to extract keywords from documents without any training data.

Model stacking is an ensemble technique that involves training a model to combine the outputs of many diverse statistical models, and has been shown to improve predictive performance in a variety of settings. stacks implements a grammar for tidymodels'-aligned model stacking.

This package provides easy to use functions to create all-sky grid plots of widely used astronomical coordinate systems (equatorial, ecliptic, galactic) and scatter plots of data on any of these systems including on-the-fly system conversion. It supports any type of spherical projection to the plane defined by the mapproj package.

This package provides a collection of functions to perform Detrended Fluctuation Analysis (DFA exponent), GUEDES et al. (2019) <doi:10.1016/j.physa.2019.04.132> , Detrended cross-correlation coefficient (RHODCCA), GUEDES & ZEBENDE (2019) <doi:10.1016/j.physa.2019.121286>, DMCA cross-correlation coefficient and Detrended multiple cross-correlation coefficient (DMC), GUEDES & SILVA-FILHO & ZEBENDE (2018) <doi:10.1016/j.physa.2021.125990>, both with sliding windows approach.

The goal of SAFEPG is to predict climate-related extreme losses by fitting a frequency-severity model. It improves predictive performance by introducing a sign-aligned regularization term, which ensures consistent signs for the coefficients across the frequency and severity components. This enhancement not only increases model accuracy but also enhances its interpretability, making it more suitable for practical applications in risk assessment.

Input/Output, processing and visualization of spectra taken with different spectrometers, including SVC (Spectra Vista), ASD and PSR (Spectral Evolution). Implements an S3 class spectra that other packages can build on. Provides methods to access, plot, manipulate, splice sensor overlap, vector normalize and smooth spectra.

This package provides a comprehensive statistical toolbox for National Statistics Institutes (INS) in Africa. Provides functions for survey data import ('KoboToolbox', ODK', CSPro', Excel', Stata', SPSS'), data processing and validation, weighted statistical analysis (descriptive statistics, cross-tabulations, regression, Human Development Index (HDI), Multidimensional Poverty Index (MPI) following Alkire and Foster (2011) <doi:10.1093/oep/gpr051>, inequalities), visualization (age pyramids, thematic maps, official charts) and dissemination ('SDMX export, DDI metadata, anonymization, Word/PDF reports). Designed to work in resource-constrained environments, offline and in French.

This package provides a collection of statistical and geometrical tools including the aligned rank transform (ART; Higgins et al. 1990 <doi:10.4148/2475-7772.1443>; Peterson 2002 <doi:10.22237/jmasm/1020255240>; Wobbrock et al. 2011 <doi:10.1145/1978942.1978963>), 2-D histograms and histograms with overlapping bins, a function for making all possible formulae within a set of constraints, amongst others.

The complete scripts from the American version of the Office television show in tibble format. Use this package to analyze and have fun with text from the best series of all time.

This package provides functions for obtaining p-values (for hypothesis tests), confidence intervals, and multivariate confidence sets. In particular, the method is compatible with differentially private dataset, as long as the privacy mechanism is known. For more details, see Awan and Wang (2024), "Simulation-based, Finite-sample Inference for Privatized Data", <doi:10.48550/arXiv.2303.05328>.

Evaluates the statistical coherence of existing administrative partitions (e.g. inter-municipal groupings, districts) by identifying spatial units whose attributive profile is more similar to a neighbouring group than to their own. Border units are iteratively reassigned to the group they are most affine with, based on Euclidean or Mahalanobis distance computed on user-supplied numeric variables, with optional per-variable weighting and standardisation. Spatial contiguity is enforced throughout: isolated candidates are reintegrated into their original group, disconnected fragments are resolved, and empty groups are restored. Convergence is monitored via an eta-squared cohesion criterion. The resulting partition can be compared to the original administrative delineation using multilevel models, providing a quantitative measure of boundary inefficiency.

Integrates the 13C nuclear magnetic resonance spectra using different integration ranges. Output depends on the method chosen. For the Molecular Mixing Model, a measurement of the fitting quality is given by its R-factor. For more details see: <doi:10.5281/zenodo.10137768>.

This package provides an R interface to the sparseLM C library for large-scale nonlinear least squares problems with arbitrarily sparse Jacobians. The underlying solver implements a sparse variant of the Levenberg-Marquardt algorithm for minimizing sum-of-squares objective functions, supports user-supplied analytic Jacobians or finite-difference approximation, and is designed to exploit sparsity for improved memory use and performance. This package exposes the solver in R and uses sparse matrix classes and the CHOLMOD sparse Cholesky factorization routines through the Matrix package interface. Methods from the C library are described in Lourakis (2010) <doi:10.1007/978-3-642-15552-9_4>.