# November 2019: "Top 40" New R Packages

by Joseph Rickert

One hundred forty-four new packages made it to CRAN in November. Here are my picks for the “Top 40” in eight categories: Computational Methods, Data, Genomics, Machine Learning, Statistics, Time Series, Utilities, and Visualization.

### Computational Methods

calculus v0.1.1: Provides C++ optimized functions for numerical and symbolic calculus including symbolic arithmetic, tensor calculus, Einstein summation convention, Taylor series expansion, multivariate Hermite polynomials and much more.

Jaya v0.1.9: Implements a gradient-free algorithm, without hyperparameters, for solving both constrained and unconstrained optimization problems. See Rao (2016) for details and the vignette for how to use the package.

treenomial v1.1.1: Provides functions for creating and comparing polynomials that uniquely describe trees as introduced in Liu (2019). See README for information.

### Data

eudract v0.9.0: Provides access to the European Clinical Trials Data Base ( EudraCT), which summarizes of all registered clinical trial results. The intent is to prevent non-reporting of negative results and provide open-access to results to inform future research. There is a vignette.

ozmaps v0.2.0: Provides maps of Australian coastline and administrative regions as well as simple functions for country or state maps of Australia, and in-built data sets of administrative regions from the Australian Bureau of Statistics. See README for examples.

presentes v0.1.0: Provides a compilation and digitization of the official registry of victims of state terrorism in Argentina during the last military coup. The original data comes from RUVTE-ILID (2019) research and the List of Victims.

VancouvR : Provides a wrapper for the City of Vancouver Open Data API. There is an Introduction.

wiesbaden v1.2.0: Implements an interface to retrieve and import data from different databases of the Federal Statistical Office of Germany (DESTATIS). There is a vignette.

### Genomics

biocompute v1.0.3: Provides tools to create, validate, and export BioCompute Objects as described in King et al. (2019). There is an Introduction and a vignette on Authoring Biocompute Objects.

diem v1.0: Implements a novel semi-supervised machine learning classier DIEM, Debris Identification using Expectation Maximization, to identify debris-containing droplets from a droplet-based single cell/nucleus RNA-seq. See the vignette for details.

### Machine Learning

azuremlsdk v0.5.7: Implements an interface to the Azure Machine Learning Software Development Kit enabling data scientists to train, deploy, automate, and manage machine learning models on the Azure Machine Learning service. There are vignettes on Setup and Installation.

hereR v0.2.0: Implements an interface to the HERE REST APIs which provide information on geocoding, routing directions, traffic flow, and weather forecasts. There are vignettes on Authentication, the Geocoder API, the Routing API, the Traffic API, and the Weather API.

hilbertSimilarity v0.4.3: Uses Hilbert Curves to develop the notion of Hilbert Similarity to quantify the similarity between samples in high dimensional data. There are vignettes on Comparing Samples and Identifying Treatment Effects.

orf v0.1.2: Implements the Ordered Forest estimator as developed in Lechner & Okasa (2019) to estimate the conditional probabilities of models with ordered categorical outcome (ordered choice models). See the vignette for details.

RPEClust v0.1.0: Implements the random projection ensemble clustering algorithm described in Anderlucci et al.(2019) and Raftery and Dean (2006).

### Statistics

DiffXTables v0.0.2: Provides functions for statistical hypothesis testing of pattern heterogeneity via differences in underlying distributions across two or more contingency tables. It includes the comparative chi-squared test, the Sharma-Song test, and the heterogeneity test. See the vignette for details.

effectsize v0.0.1: Provides functions to work with indices of effect size and standardized parameters for a wide variety of models (See Lüdecke, Waggoner & Makowski (2019).) There are vignettes on Bayesian Models, Converting Between Incices, Automated Interpretation of Indices, Data Standardization and Parameter Standardization.

exPrior v1.0.1: Provides practitioners of statistics in geology, hydrology, etc. with a tool for deriving prior distributions for Bayesian inference. See the vignette.

fitHeavyTail v0.1.1: Implements robust estimation methods for the mean vector and covariance matrix from data (possibly containing NAs) under multivariate heavy-tailed distributions such as angular Gaussian, Cauchy, and Student’s t. See Sun et al. (2014), Sun et al. (2015), Liu and Rubin (1995) and Zhou et al. (2015) for background, and the vignette for examples.

mixl v1.1: Provides functions for simulated maximum likelihood estimation of multinomial logit models, mixed models, random coefficients and hybrid choice models. See Molloy et al. (2019) for details and the User Guide.

MKdescr v0.5: Provides functions to compute a standardized interquartile range (IQR), a Huber-type skipped mean as described in Hampel (1985), a robust coefficient of variation as described in Arachchige et al. (2019), a robust signal to noise ratio (SNR), and more. See the vignette.

MNLpred v0.0.1: Provides functions to return simulated predicted probabilities and first differences for multinomial logit models. The methodological approach is based on the principles laid out by King, Tomz, and Wittenberg (2000). See the vignette for examples.

pdqr v0.2.0: Provides functions to create, transform, and summarize custom discrete and continuous random variables with distribution functions that are analogues of p*(), d*(), q*(), and r*(). There are vignettes on Creating, Converting, Transforming, and Summarizing pdqr functions.

tensorregression v1.0: Implements the generalized tensor regression in Xu, Hu and Wang (2019) to solve tensor-response regression given covariates on multiple modes with alternating updating algorithm.

tidydice: v0.0.4: Provides functions for basic statistical experiments, that can be used for teaching introductory statistics. See the vignette.

tvgeom v1.0.1: Implements the probability mass, distribution, quantile, and random number generating functions for the time-varying right-truncated geometric distribution. See the vignette for background.

### Time Series

gravitas v0.1.0: Provides tools for systematically exploring large quantities of temporal data across different temporal granularities (deconstructions of time) by visualizing probability distributions. There are vignettes on exploring probability distributions for cricket and for bivariate temporal franularities.

smoots v1.0.1: Provides nonparametric estimates of trend and its derivatives in equidistant time series with short-memory stationary errors. See Feng and Gries (2017) for the methods employed, and see README for an example.

tsfgrnn v0.1.0: Implements a general regression neural network (GRNN), a variant of a radial basis function network, for forecasting time series. See Martinez et al. (2019) and Yan (2012) for background and the vignette for examples.

### Utilities

dipsaus v0.0.3: Provides enhancement functions that fall into four categories: shiny input widgets; high-performance computing using RcppParallel and future; functions to modify R calls and convert numbers, strings, and other objects; and utility functions to get system information such as CPU chipset, memory limit, etc. See the vignettes: Asynchronous Evaluator, R Expression Add-ons, Shiny Customized Widgets, and Utility Functions.

extraoperators v0.1.1: Provides operator functions for common tasks such as logical or relational comparisons, finding indices and subsetting. See the vignette for details.

gluedown v1.0.1: Provides functions to transition between R vectors and markdown text. Users can create vectors in R, glue strings together with the markdown syntax, and print formatted vectors directly to the document. This package primarily uses GitHub Flavored Markdown. There is a vignette on GitHub Flavored Markdown and another on Printing Markdown.

googlesheets4 v0.1.0: Provides functions for interacting with Google Sheets through the Sheets API v4. See README for help.

hdd v0.1.0: Provides a data class for importing and manipulating out of memory data sets. See the Introduction.

RVerbalExpressions v0.1.0: Provides tools to build regular expressions using grammar and functionality inspired by VerbalExpressions. See the vignette for examples.

shinyMobile v0.1.0: Provides tools for building shiny apps for iOS, Android, and desktop computers as well as beautiful shiny gadgets. shinyMobile is built on top of the latest ‘Framework7’ template. There is a Getting Started Guide and vignettes on Dark-Theme, Gadgets, Single-Layout, Split-Layout, Tabs-Layout, and Tools.

tidycwl v1.0.4: Implements the Common Workflow Language for describing data analysis workflows. See the vignette for details.

### Visualization

barplot3d v1.0.1: Provides functions for creating 3D plots including sequence context plots used in DNA sequencing analysis. See the vignette for examples.

fplot v0.2.0: Provides functions to plot regular/weighted/conditional distributions by using formulas. See the vignette for examples.

robvis: v0.3.0: Provides functions for visualizing risk-of-bias assessments performed as part of a systematic review, providing tools for randomized controlled trials ( Sterne et al. (2019)), non-randomized studies of interventions ( Sterne et al (2016)), and diagnostic accuracy studies ( Whiting et al (2011)). There is a vignette.