Efficient Inference on High-Dimensional Linear Models With Missing Outcomes

This package implements the proposed debiasing method for conducting valid inference on the high-dimensional linear regression function with missing outcomes. We also document all the code for the simulations and real-world applications in our paper here.

• Free software: MIT license
• Python Package Documentation: https://debias-infer.readthedocs.io.
• You may also consider using our R package DebiasInfer, though the Python package will be computationally faster.

Installation guide

Debias-Infer requires Python 3.8+ (earlier version might be applicable), NumPy, SciPy, scikit-learn, CVXPY, statsmodels. To install the latest version of Debias-Infer from this repository, run:

python setup.py install

To pip install a stable release, run:

pip install Debias-Infer

References

[1] Y. Zhang, A. Giessing, Y.-C. Chen (2023+) Efficient Inference on High-Dimensional Linear Models with Missing Outcomes arXiv:2309.06429.

[2] T. Sun and C.-H. Zhang (2012). Scaled Sparse Linear Regression. Biometrika, 99, no.4: 879-898.