Multiple Imputation with Denoising Autoencoders
Reverse numeric conversion of binary vector
Apply MAR missingness to data
Coalesce one-hot encoding back to a single variable
Scale numeric vector between 0 and 1
Estimate and combine regression models from multiply-imputed data
Impute missing values using imputation model
Pre-process data for Midas imputation
Delete the rMIDAS Environment and Configuration
Instantiate Midas class
Configure python for MIDAS imputation
Manually set up Python connection
Replace NA missing values with NaN
Perform overimputation diagnostic test
Check whether Python is capable of executing example code
Initialise connection to Python
Reset the rMIDAS Environment Configuration
Manually select python binary
Skip test where 'numpy' not available.
Train an imputation model using Midas
Reverse minmax scaling of numeric vector
A tool for multiply imputing missing data using 'MIDAS', a deep learning method based on denoising autoencoder neural networks (see Lall and Robinson, 2022; <doi:10.1017/pan.2020.49>). This algorithm offers significant accuracy and efficiency advantages over other multiple imputation strategies, particularly when applied to large datasets with complex features. Alongside interfacing with 'Python' to run the core algorithm, this package contains functions for processing data before and after model training, running imputation model diagnostics, generating multiple completed datasets, and estimating regression models on these datasets. For more information see Lall and Robinson (2023) <doi:10.18637/jss.v107.i09>.