Skip to content

Latest commit

 

History

History
150 lines (114 loc) · 5 KB

README.md

File metadata and controls

150 lines (114 loc) · 5 KB
Raw

interflex

Lifecycle: experimental License: MIT downloads: CRAN

interflex estimates, interprets, and visualizes marginal effects and performs diagnostics.

Authors: Jens Hainmueller, Jiehan Liu, Licheng Liu, Ziyi Liu, Jonathan Mummolo, Tianzhu Qin, and Yiqing Xu (maintainer)

Date: August 2, 2024

Repos: Github (1.3.1) CRAN (1.2.6)

Tutoralis: See tutorials for cases with continuous and discrete outcomes, as well as a tutorial for double/debiased machine learning (DML) estimators.

Examples: R code used in the tutorials can be downloaded from here.

Reference: How Much Should We Trust Estimates from Multiplicative Interaction Models? Simple Tools to Improve Empirical Practice. Political Analysis, Vol. 27, Iss. 2, April 2019, pp. 163–192.

Installation

You can install the interflex package from CRAN:

install.packages('interflex', type = "source", 
                 repos = 'http://cran.us.r-project.org')

Or you can install the up-to-date development version from Github:

# if not already installed
install.packages('devtools', repos = 'http://cran.us.r-project.org') 
# install from github
devtools::install_github('xuyiqing/interflex')

interflex depends on the following packages, which will be installed automatically when interflex is being installed; if not, please install them manually:

# Function to install packages
install_all <- function(packages) {
  installed_pkgs <- installed.packages()[, "Package"]
  for (pkg in packages) {
    if (!pkg %in% installed_pkgs) {
      install.packages(pkg, repos = 'http://cran.us.r-project.org')
    }
  }
}

# Packages to be installed
packages <- c("Rcpp", "mgcv", "sandwich", "pcse", "fixest", "foreach", "doParallel", 
              "lfe", "lmtest", "Lmoments", "ggplot2", "plotrix", "grid", 
              "gridExtra", "ggplotify", "ggpubr", "RColorBrewer", "grDevices", 
              "gtable", "MASS", "mvtnorm", "pROC", "ModelMetrics", "foreign",
              "patchwork", "rmarkdown")

# Install the packages
install_all(packages)

Mac users who encounter “-lgfortran” or “-lquadmath” error during installation, please check out the solution here. Typing the following two lines of code in your Terminal should solve this problem.

curl -OL http://r.research.att.com/libs/gfortran-4.8.2-darwin13.tar.bz2
sudo tar fvxz gfortran-4.8.2-darwin13.tar.bz2 -C /

Mac users who encounter clang: error: unsupported option ‘-fopenmp’, please consider (1) updating your R and/or (2) installing new R macro tools from Github.

Python Environment

To use the double/debiased machine learning estimators in interflex, we rely on packages in Python. The integration between R and Python is facilitated by the reticulate package, which allows R to interface with Python seamlessly. The following steps set up a Python environment with all required dependencies.

First, install the reticulate package in R. This package enables R to interface with Python.

install.packages("reticulate", repos = 'http://cran.us.r-project.org')

Then, we use Miniconda, a minimal installer for Conda. Miniconda helps manage Python environments and packages. By default, the installation process will create a virtual environment named r-reticulate.

reticulate::install_miniconda(update = TRUE)

With Miniconda installed, we point the platform using use_condaenv. This step is incorporated in the interflex package.

reticulate::use_condaenv(condaenv = "r-reticulate")

Finally, install the necessary Python libraries. These libraries include tools for statistical modeling and machine learning that are essential for using the interflex package.

reticulate::py_install(packages = c("patsy", "numpy", "pandas", 
                    "scikit-learn==1.2.0", "doubleml", "econml"))

Report bugs

Please report bugs to yiqingxu [at] stanford.edu with your sample code and data file. Much appreciated!