General P-splines are non-uniform B-splines penalized by a general difference penalty, proposed by Li and Cao (2022) <arXiv:2201.06808>. Constructible on arbitrary knots, they extend the standard P-splines of Eilers and Marx (1996) <doi:10.1214/ss/1038425655>. They are also related to the O-splines of O'Sullivan (1986) <doi:10.1214/ss/1177013525> via a sandwich formula that links a general difference penalty to a derivative penalty. The package includes routines for setting up and handling difference and derivative penalties. It also fits P-splines and O-splines to (x, y) data (optionally weighted) for a grid of smoothing parameter values in the automatic search intervals of Li and Cao (2023) <doi:10.1007/s11222-022-10178-z>. It aims to facilitate other packages to implement P-splines or O-splines as a smoothing tool in their model estimation framework.

We provides functions that employ splines to estimate generalized partially linear single index models (GPLSIM), which extend the generalized linear models to include nonlinear effect for some predictors. Please see Y. (2017) at <doi:10.1007/s11222-016-9639-0> and Y., and R. (2002) at <doi:10.1198/016214502388618861> for more details.

Fits generalized linear models where the parameters are subject to linear constraints. The model is specified by giving a symbolic description of the linear predictor, a description of the error distribution, and a matrix of constraints on the parameters.

This package performs binary classification via Group Method of Data Handling (GMDH) - type neural network algorithms. There exist two main algorithms available in GMDH() and dceGMDH() functions. GMDH() performs classification via GMDH algorithm for a binary response and returns important variables. dceGMDH() performs classification via diverse classifiers ensemble based on GMDH (dce-GMDH) algorithm. Also, the package produces a well-formatted table of descriptives for a binary response. Moreover, it produces confusion matrix, its related statistics and scatter plot (2D and 3D) with classification labels of binary classes to assess the prediction performance. All GMDH2 functions are designed for a binary response (Dag et al., 2019, <https://download.atlantis-press.com/article/125911202.pdf>).

Make 2D and 3D plots of linear programming (LP), integer linear programming (ILP), or mixed integer linear programming (MILP) models with up to three objectives. Plots of both the solution and criterion space are possible. For instance the non-dominated (Pareto) set for bi-objective LP/ILP/MILP programming models (see vignettes for an overview). The package also contains an function for checking if a point is inside the convex hull.

This package implements a variant of the Self-Organizing Map (SOM) algorithm designed for mixed-attribute datasets. Similarity between observations is computed using the Gower distance, and categorical prototypes are updated via heuristic strategies (weighted mode and multinomial sampling). Provides functions for model fitting, mapping, visualization (U-Matrix and component planes), and evaluation, making SOM applicable to heterogeneous real-world data. For methodological details see Sáez and Salas (2026) <doi:10.1007/s41060-025-00941-6>.

Causal mediation analysis for a single exposure/treatment and a single mediator, both allowed to be either continuous or binary. The package implements the difference method and provides point and interval estimates as well as testing for the natural direct and indirect effects and the mediation proportion. Nevo, Xiao and Spiegelman (2017) <doi:10.1515/ijb-2017-0006>.

Conducts causal inference with interactive fixed-effect models. It imputes counterfactuals for each treated unit using control group information based on a linear interactive fixed effects model that incorporates unit-specific intercepts interacted with time-varying coefficients. This method generalizes the synthetic control method to the case of multiple treated units and variable treatment periods, and improves efficiency and interpretability.

Generate multiple data sets for educational purposes to demonstrate the importance of multiple regression. The genset function generates a data set from an initial data set to have the same summary statistics (mean, median, and standard deviation) but opposing regression results.

Calculate, plot and animate the configuration of Jupiter's four largest satellites (known as Galilean satellites) for a given date and time (UTC - Coordinated Universal Time). The galsat() function returns numerical values of the satellitesâ positions. x â the apparent rectangular coordinate of the satellite with respect to the center of Jupiterâ s disk in the equatorial plane in the units of Jupiterâ s equatorial radius; X is positive toward the west, y â the apparent rectangular coordinate of the satellite with respect to the center of Jupiterâ s disk from the equatorial plane in the units of Jupiterâ s equatorial radius; Y is positive toward the north. For more details see Meeus (1988, ISBN 0-943396-22-0) "Astronomical Formulae for Calculators". The galsat_animate() function creates an animation of the Galilean satellites positions. You provide the starting time, duration, the time step between frames, and the pause between frames. The function delta_t() returns the value of delta-T in units of seconds.

Images are provided as an array dataset of 2D image thumbnails from Google Image Search <https://www.google.com/search>. This array data may be suitable for a training data of machine learning or deep learning as a first trial.

This package provides a GraphQL client, with an R6 interface for initializing a connection to a GraphQL instance, and methods for constructing queries, including fragments and parameterized queries. Queries are checked with the libgraphqlparser C++ parser via the graphql package.

In computationally demanding data analysis pipelines, the targets R package (2021, <doi:10.21105/joss.02959>) maintains an up-to-date set of results while skipping tasks that do not need to rerun. This process increases speed and increases trust in the final end product. However, it also overwrites old output with new output, and past results disappear by default. To preserve historical output, the gittargets package captures version-controlled snapshots of the data store, and each snapshot links to the underlying commit of the source code. That way, when the user rolls back the code to a previous branch or commit, gittargets can recover the data contemporaneous with that commit so that all targets remain up to date.

Encodes simple feature ('sf') objects and coordinates, and decodes polylines using the Google polyline encoding algorithm (<https://developers.google.com/maps/documentation/utilities/polylinealgorithm>).

This package provides methods for searching through genealogical data and displaying the results. Plotting algorithms assist with data exploration and publication-quality image generation. Includes interactive genealogy visualization tools. Provides parsing and calculation methods for variables in descendant branches of interest. Uses the Grammar of Graphics.

Fits unimodal and multimodal gambin distributions to species-abundance distributions from ecological data, as in in Matthews et al. (2014) <DOI:10.1111/ecog.00861>. gambin is short for gamma-binomial'. The main function is fit_abundances(), which estimates the alpha parameter(s) of the gambin distribution using maximum likelihood. Functions are also provided to generate the gambin distribution and for calculating likelihood statistics.

This package provides a high performance interface to the Global Biodiversity Information Facility, GBIF'. In contrast to rgbif', which can access small subsets of GBIF data through web-based queries to a central server, gbifdb provides enhanced performance for R users performing large-scale analyses on servers and cloud computing providers, providing full support for arbitrary SQL or dplyr operations on the complete GBIF data tables (now over 1 billion records, and over a terabyte in size). gbifdb accesses a copy of the GBIF data in parquet format, which is already readily available in commercial computing clouds such as the Amazon Open Data portal and the Microsoft Planetary Computer, or can be accessed directly without downloading, or downloaded to any server with suitable bandwidth and storage space. The high-performance techniques for local and remote access are described in <https://duckdb.org/why_duckdb> and <https://arrow.apache.org/docs/r/articles/fs.html> respectively.

These Bayesian models written in the Stan probabilistic language can be used to interpret green crab trapping and environmental DNA monitoring data, either independently or jointly. Detailed model information is found in Keller (2022) <doi:10.1002/eap.2561>.

Light procedures for learning Global Vector Autoregression model (GVAR) of Pesaran, Schuermann and Weiner (2004) <DOI:10.1198/073500104000000019> and Dees, di Mauro, Pesaran and Smith (2007) <DOI:10.1002/jae.932>.

Moon charts are like pie charts except that the proportions are shown as crescent or gibbous portions of a circle, like the lit and unlit portions of the moon. As such, they work best with only one or two groups. gggibbous extends ggplot2 to allow for plotting multiple moon charts in a single panel and does not require a square coordinate system.

This package provides functions for fitting and doing predictions with Gaussian process models using Vecchia's (1988) approximation. Package also includes functions for reordering input locations, finding ordered nearest neighbors (with help from FNN package), grouping operations, and conditional simulations. Covariance functions for spatial and spatial-temporal data on Euclidean domains and spheres are provided. The original approximation is due to Vecchia (1988) <http://www.jstor.org/stable/2345768>, and the reordering and grouping methods are from Guinness (2018) <doi:10.1080/00401706.2018.1437476>. Model fitting employs a Fisher scoring algorithm described in Guinness (2019) <doi:10.48550/arXiv.1905.08374>.

Extended techniques for generalized linear models (GLMs), especially for binary responses, including parametric links and heteroscedastic latent variables.

Easily explore data by plotting graphs with a few lines of code. Use these ggplot() wrappers to quickly draw graphs of scatter/dots with box-whiskers, violins or SD error bars, data distributions, before-after graphs, factorial ANOVA and more. Customise graphs in many ways, for example, by choosing from colour blind-friendly palettes (12 discreet, 3 continuous and 2 divergent palettes). Use the simple code for ANOVA as ordinary (lm()) or mixed-effects linear models (lmer()), including randomised-block or repeated-measures designs, and fit non-linear outcomes as a generalised additive model (gam) using mgcv(). Obtain estimated marginal means and perform post-hoc comparisons on fitted models (via emmeans()). Also includes small datasets for practising code and teaching basics before users move on to more complex designs. See vignettes for details on usage <https://grafify.shenoylab.com/>. Citation: <doi:10.5281/zenodo.5136508>.

Unconstrained and constrained maximum likelihood estimation of structural and reduced form Gaussian mixture vector autoregressive, Student's t mixture vector autoregressive, and Gaussian and Student's t mixture vector autoregressive models, quantile residual tests, graphical diagnostics, simulations, forecasting, and estimation of generalized impulse response function and generalized forecast error variance decomposition. Leena Kalliovirta, Mika Meitz, Pentti Saikkonen (2016) <doi:10.1016/j.jeconom.2016.02.012>, Savi Virolainen (2025) <doi:10.1080/07350015.2024.2322090>, Savi Virolainen (in press) <doi:10.1016/j.ecosta.2025.09.003>.