MIDASpy is a Python package for multiply imputing missing data using deep learning methods. The MIDASpy algorithm offers significant accuracy and efficiency advantages over other multiple imputation strategies, particularly when applied to large datasets with complex features. In addition to implementing the algorithm, the 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.

Background and suggested citation

For more information on MIDAS, the method underlying the software, see:

Lall, Ranjit, and Thomas Robinson. Forthcoming. “The MIDAS Touch: Accurate and Scalable Missing-Data Imputation with Deep Learning.” Political Analysis. Gated. Ungated.


To install via pip, enter the following command into the terminal:
pip install MIDASpy

The latest development version (potentially unstable) can be installed
via the terminal with: pip install git+https://github.com/MIDASverse/MIDASpy.git

MIDAS requires:

  • Python (>=3.5; <3.9)
  • Numpy (>=1.5)
  • Pandas (>=0.19)
  • Tensorflow (>= 1.10) – TensorFlow>=2.2 now fully supported
  • Matplotlib
  • Statmodels
  • Scipy
  • TensorFlow Addons (>=0.11 if using Tensorflow >= 2.2)

Tensorflow also has a number of requirements, particularly if GPU
acceleration is desired. See https://www.tensorflow.org/install/ for


For a simple demonstration of MIDASpy, see our Jupyter Notebook

Version 1.2.1 (January 2021)

v1.2.1 adds new pre-processing functionality and a multiple imputation
regression function.

Users can now automatically preprocess binary and categorical columns
prior to running the MIDAS algorithm using binary_conv() and

The new combine() function allows users to run regression analysis
across the complete data, following Rubin’s combination rules.

Version 1.1.1 (October 2020)

v1.1.1 fixes a minor dependency bug.

Update adds full Tensorflow 2.X support:

  • Users can now run the MIDAS algorithm in TensorFlow 2.X (TF1 support

  • Tidier handling of random seed setting across both TensorFlow and

  • Minor bug fixes

Previous versions

Version 1.0.2 (September 2020)

Key changes:

  • Minor, mainly cosmetic, changes to the underlying source code.
  • Renamed ‘categorical_columns’ argument in build_model() to
    ‘binary_columns’ to avoid confusion
  • Added plotting arguments to overimputation() method to suppress
    intermediary overimputation plots (plot_main) and all plots
  • Changed overimputation() plot titles, labels and legends
  • Added tensorflow 2.0 version check on import
  • Fixed seed-setting bug in earlier versions

Alpha 0.2:

Variational autoencoder enabled. More flexibility in model
specification, although defaulting to a simple mirrored system. Deeper
analysis tools within .overimpute() for checking fit on continuous
values. Constructor code deconflicted. Individual output specification
enabled for very large datasets.

Key added features:

  • Variational autoencoder capacity added, including encoding to and
    sampling from latent space

Planned features:

  • Time dependence handling through recurrent cells
  • Improving the pipeline methods for very large datasets
  • Tensorboard integration
  • Dropout scaling
  • A modified constructor that can generate embeddings for better
    interpolation of features
  • R support

Wish list:

  • Smoothing for time series (LOESS?)
  • Informative priors?