Two hundred ninety-six new packages made it to CRAN in March. Here are my “Top 40” picks in ten categories: Computational Methods, Data, Machine Learning, Mathematics, Medicine, Science, Statistics, Time Series, Utilities, and Visualization.
celltrackR v0.3.1: Provides a methodology to analyze cells that move in a two- or three-dimensional space. While the methodology has been developed for cell trajectory analysis, it is applicable to anything that moves including animals, people, or vehicles. For background see Beauchemin et al. (2007), Beltman et al. (2009), Gneiting & Schlather (2004) and additional papers listed by the package authors. There are There is an Introduction and vignettes on Track Analysis Methods, Clustering, Quality Control and Preprocessing, and Simulating Tracks.
collapse v1.1.0: Implements C/C++ based functions for advanced data transformations including statistical functions supporting grouped and/or weighted computations on vectors, matrices and data.frames and more. There is an Introduction and vignettes on collapse and dplyr, and Collapse and plm.
graphsim v0.1.1: Provides functions to simulate continuous data (e.g., gene expression) from a sigma covariance matrix derived from a graph structure in
igraph objects. It extends
mvtnorm to take
igraph structures as input. There are vignettes on simulating gene expression data, convergent graph structures, divergent graph structures, network graph structures, inhibiting network graph structures, reconvergent graph structures, and vignettes on Directed Plots and Viral Plots.
rkeops Provides an interface to the KeOps library which computes generic reductions of very large arrays whose entries are given by a mathematical formula, and is suited to the computation of kernel dot products and the associated gradients, even when the full kernel matrix does not fit into the GPU memory. There is an Introduction and a vignette on Bindings.
simhelpers v0.1.0: Provides functions to help run simulation studies and calculate performance measures and associated Monte Carlo standard errors for simulation results. The general simulation workflow is closely aligned with the approach described by Morris et al. (2019). There are vignettes on Performance Criteria, Workflow, and Visualization.
COVID19 v1.0.0: Provides COVID-19 datasets from several sources in a unified tidy format. The data are downloaded in real-time, cleaned and matched with exogenous variables. Vintage databases are also supported. There is a vignette.
ReDaMoR v).4.2: Implements methods to manipulate relational data models, including functions to create, modify and export data models in
json format, importing models created with MySQL Workbench, and a shiny app. See the Tutorial.
SSLR v0.9.1: Implements techniques for semi-supervised (both labeled and unlabeled data are used to train a classifier) classification and regression. There is an [Introduction]() and vignettes on Model Fitting, Models, Classification and Regression.
actuaryr v1.1.1: Provides functions to refer to the first or last (working) day within a specific period relative to a base date to facilitate actuarial reporting and to compare results. See the vignette for details.
stokes v1.0-5: Provides functionality for working with differentials, k-forms, wedge products, Stokes’s theorem, and related concepts from the exterior calculus and also Grassman algebra. See Calculus on Manifolds for the math, and the vignette for an introduction.
escalation v0.1.2: Implements methods for working with dose-finding clinical trials and includes a common interface to various dose-finding methodologies such as the continual reassessment method (CRM) by O’Quigley et al. (1990), the Bayesian optimal interval design (BOIN) by Liu & Yuan (2015), and the 3+3 described by Korn et al. (1994). There are vignettes on Working with dose-paths, Working with dose selectors, and Simulating dose-escalation trials.
AQuadtree v1.0.0: Implements an automatic aggregation tool to manage point data privacy using the methodology described in Lagonigro et al. (2017). The algorithm seeks data accuracy at the smallest possible areas preventing individual information disclosure. The vignette provides examples.
bbricks v0.1.2: Provides tools to fit Bayesian parametric and nonparametric models including Gaussian and Normal-Inverse-Wishart conjugate structure, Gaussian and Normal-Inverse-Gamma conjugate structure, Categorical and Dirichlet conjugate structure and other Dirichlet processes. See the vignette.
causaloptim v0.6.5: Implements an interface to specify causal graphs and compute bounds on causal effects by extending and generalizing the approach taken in Balke & Pearl (1994). There are vignettes on Examples, Computing Causal Bounds, and on the Using Shiny App.
detectseparation v0.1: Provides pre-fit and post-fit methods for detecting separation and infinite maximum likelihood estimates in generalized linear models with categorical responses. The pre-fit methods solve the linear programming problems for the detection of separation developed in Konis (2007) The post-fit methods apply to models with categorical responses, including binomial-response generalized linear models and multinomial-response models. See the vignette for more information.
multilevelTools v0.1.1: Computes effect sizes, diagnostics and performance metrics for multilevel and mixed effects models including marginal and conditional R2 estimates for linear mixed effects models based on Johnson (2014). See the vignette.
registr v1.0.0: Implements the functional data analysis method for registering curves (functional data), described in Wrobel et al. (2019) that are generated from exponential family distributions. The vignette provides an introduction.
specr v0.2.1: Provides utilities for conducting specification curve analyses, Simonsohn et al. (2015), or multiverse analyses ( Steegen et al. (2016)) including functions to setup, run, evaluate, and plot all specifications. There is an Introduction and vignettes on Customizing Plots, Identifying Variance Components, Investigating Selected Specifications, and Visualizing Progress.
spsurv v1.0.0: Provides routines to ease semiparametric survival regression modeling based on Bernstein polynomials, including proportional hazards, proportional odds and accelerated failure time frameworks for right-censored data. See Panaro (2020) for background and the vignette to get started.
terra v0.5-8: Provides classes and methods for spatial data as well as methods to allow for low-level data manipulation as well as high-level global, local, zonal, and focal computation. See the manual and tutorials to get started.
tramME v0.0.2: Implements likelihood-based estimation of mixed-effects transformation models using the Template Model Builder. See Hothorn et al. (2018) for the technical details and the vignette to get started.
fsMTS v0.1.5: Implements feature selection routines for multivariate time series including external structure, Pfeifer & Deutsch (1980); cross-correlation; graphical LASSO, Haworth & Cheng (2014), least angle regression, Gelper & Croux (2008); mutual information Liu et al. (2016), and partial spectral coherence Davis et al. (2016). There is a vignette using simulated data and another using real traffic data.
brio v1.0.0: Provides functions to handle basic input output, these functions always read and write UTF-8 (8-bit Unicode Transformation Format) files and provide more explicit control over line endings.
dm v0.1.1: Provides tools for working with multiple related tables, stored as data frames or in a relational database. There is a Getting Started Guide, and vignettes on Basic Operations, Filtering, Function Naming Logic, Relational Data Models, Joining, Low Level Operations. Data Prep, Databases, Visualizing
dm Objects, and on Zooming and Manipulating tables.
ipaddress v0.2.0: Provides classes and functions for working with IP (Internet Protocol) addresses and networks and offers full support for both IPv4 and IPv6 (Internet Protocol versions 4 and 6) address spaces. There is an Introduction.
scaffolder v0.0.1: Comprehensive set of tools for scaffolding R interfaces to modules, classes, functions, and documentations written in other programming languages, such as
Python. See the vignette for examples.
colorBlindness v0.1.6: Provide the safe color sets for color blindness collected from: Wong, B. (2011), bcgs and University of Oregon and also simulators of protanopia, deuteranopia based on Vienot et al. (1999). See the vignette for examples.