In the course of my research projects, I have found it useful to develop a few R packages, which can all be found on my GitHub. A few are in mature stages of development:
RcppDist provides header-only C++ library with functions for additional statistical distributions that can be called from C++ when writing code using Rcpp or RcppArmadillo.
I had occasion to draw up the C++ code for these distributions on various projects, and rolled them all together into an R package and got it up on CRAN so no one else would have to.
bggum provides R tools for a Bayesian approach to estimating Generalized Graded Unfolding Model (GGUM) (Roberts, Donoghue, and Laughlin 2000) parameters.
See the description of “Ends Against the Middle: Scaling Votes when Ideological Opposites Behave the Same for Antithetical Reasons,” my paper with Jacob Montgomery, on my research page for more information.
gpirt provides a C++ implementation with R bindings of the Gaussian process
IRT model presented in (Duck-Mayr, Garnett, and Montgomery 2020) (paper
available on my research page).
gpmss provides R tools for Gaussian process regression and classification.
While several other R packages provide some of this functionality,
this package presents a uniform user interface between models and provides some
tools not available in any other implementation, such as average marginal
effects which I derive for GP models in my dissertation’s second chapter,
“Inference in Gaussian Process Models for Political Science”,
available on my research page.