R is a highly extensible statistical programming and graphics language. R's power comes from user-contributed extensions, called packages. Packages can be uploaded to a central repository, such as CRAN, Bioconductor, or Omegahat. However, many excellent packages live out their lives in more informal code repositories such as GitHub, or even just on a single user's computer.
Packages can be peer-reviewed through many processes, such as submitting companion articles to journals like the Journal of Open Source Software or the Journal of Statistical Software. Packages can also be peer-reviewed through the rOpenSci project. I am an rOpenSci reviewer.
Here is a list of R packages that I have created or contributed to. Click on the package name to be brought to the package website.
Parse a BibTeX file to a
data.frame to make it accessible for further analysis and visualization.
Tools for exploring data from the MLA International Bibliography exported to RIS format.
Create a random monster name in the style of Diablo II.
Provides random matrices with a defined covariance structure. The package began as an exercise to show performance improvements of using vectorized R code rather than explicit
for loops. Later, it became a vehicle for working with the
RcppArmadillo packages and learning how to incorporate them into packages of my own. Future work may include
FORTRAN code to further explore the capabilities of using compiled code in R scripts and packages.
You can see an exploration the thought process just described in the package vignette. Different methods for function definitions are used and performance is assessed.
nomisr is for accessing Access UK official statistics from the Nomis database through R. Nomis contains data from the Census, the Labour Force Survey, DWP benefit statistics and other economic and demographic data, and is maintained on behalf of the Office for National Statistics by the University of Durham.
nomisr package provides functions to find what data is available, the variables and query options for different datasets and a function for downloading data.
nomisr returns data in
tibble format. Most of the data available through
nomisr is based around statistical geographies, with a handful of exceptions.
The package is for demographers, economists, geographers, public health researchers and any other researchers who are interested in geographic factors. The package aims to aid reproducibility, reduce the need to manually download area profiles, and allow easy linking of different datasets covering the same geographic area.
Split REDCap repeating instruments output into multiple tables. This will take raw output from a REDCap export and split it into a base table and child tables for each repeating instrument. This functionality was also ported into SAS!
Calculates the power and sample size for Cochran-Mantel-Haenszel tests. There are also several helper functions for working with probability, odds, relative risk, and odds ratio values. This package has been published on CRAN!
Calculate two sample Student's t-test using summary statistics
rather than feeding it raw data, such as with the
Labor force data provided by the US Bureau of Labor and Statistics. From their website (at https://www.bls.gov/lau/):
The Local Area Unemployment Statistics (LAUS) program produces monthly and annual employment, unemployment, and labor force data for Census regions and divisions, States, counties, metropolitan areas, and many cities, by place of residence.
This package is a
learnr demo built for the West Michigan R Users Group
meeting at Big Data Ignite 2017. It is meant to introduce R to a user with some background in coding but who has had little to no experience with R per se.