rMIDAS

Overview

rMIDAS is a multiple imputation package for R using deep learning, providing a simplified workflow to multiply impute and analyse data:

• convert() carries out all necessary pre-processing steps
• train() constructs and trains a MIDAS imputation model.
• complete() generates multiple completed datasets from the trained model
• combine() runs regression analysis across the complete data, following Rubin’s Rules.

rMIDAS is based on MIDASpy. More information about the underlying imputation method can be found here.

Efficient handling of large data

rMIDAS also incorporates several features to streamline and improve the efficiency of multiple imputation analysis:

• Optimisation for large datasets using data.table and mltools packages
• Automatic reversing of all pre-processing steps prior to analysis
• Built-in regression function based on glm (applying Rubin’s rules)

Installation

rMIDAS is now available on CRAN. To install the package in R, you can use the following code:

install.packages("rMIDAS")

To install the latest development version, please use the following code:

# install.packages("devtools")
devtools::install_github("MIDASverse/rMIDAS")

Note that rMIDAS uses the reticulate package to interface with Python. Users must have Python 3.X installed in order to run MIDAS. rMIDAS will automatically try to find Python 3 unless users specify their own version, using the following code:

library(rMIDAS)

# Point to a Python binary
set_python_env(path = "path/to/python/binary", type = "auto", exact = FALSE)

# Point to a virtualenv binary
set_python_env(path = "path/to/virtual/env", type = "virtualenv", exact = FALSE)

# Point to a condaenv, where conda can be supplied to choose a specific executable
set_python_env(path = "path/to/conda/env", type = "auto", exact = FALSE, conda = "auto")

Getting help

rMIDAS is still in development, and we may not have caught all bugs. If you come across any difficulties, or have any suggestions for improvements, please raise an issue here.