qacReg

The qacReg provides easy access to linear and logistic regression with enhanced performance metrics, visualizations, and diagnostics.

Installation

You can install the development version of qacReg using:

# install remotes package if necessary
if(!require(remotes)){
  install.packages("remotes")
}

# install dominanceanalysis package from CRAN archive
archive <- "https://cran.r-project.org/src/contrib/Archive"
pkg <- "dominanceanalysis/dominanceanalysis_2.0.0.tar.gz"
file <- paste(archive, pkg, sep = "/")
install.packages(file, repos = NULL)

# install qacReg package from github
remotes::install_github("rkabacoff/qacReg")

Workflow

The package supports a simple workflow embodied in 6 functions:

library(qacReg)
model <- regress(formula, data)       # fit a linear or logistic model
info(model)                           # review detailed results
diagnostics(model)                    # perform regression diagnostics
me_plots(model)                       # visualize conditional relationships
relimp(model)                         # assess variable importance
performance(model)                    # evaluate predictive performance

See the Vignettes for examples.

Why a duck?

qacReg is on of a suite of packages developed in the Quantitative Analysis Center (QAC) at Wesleyan University. QAC = QUACK. It seemed appropriate at the time.