pylmrob
==========

This project is a Python port of the `lmrob` MM-estimator and supporting
machinery from the R package `robustbase`.

Algorithms, defaults, tuning constant tables, and certain numerical
choices are derived from `robustbase`. We gratefully acknowledge the
authors and maintainers of `robustbase`:

  Martin Maechler, Peter Rousseeuw, Christophe Croux, Valentin Todorov,
  Andreas Ruckstuhl, Matias Salibian-Barrera, Tobias Verbeke,
  Manuel Koller, Eduardo L. T. Conceicao, Maria Anna di Palma.

Upstream project:
  https://cran.r-project.org/package=robustbase
  https://robustbase.r-forge.r-project.org/

Upstream license: GPL (>= 2).

This project is licensed under the GNU General Public License, version 3,
to remain compatible with `robustbase`. See LICENSE for the full text.

References
----------

The implementation follows the published algorithms:

  Yohai, V. J. (1987). High breakdown-point and high efficiency robust
  estimates for regression. Annals of Statistics 15, 642-656.

  Salibian-Barrera, M. and Yohai, V. J. (2006). A fast algorithm for
  S-regression estimates. JCGS 15, 414-427.

  Maronna, R. A. and Yohai, V. J. (2000). Robust regression with both
  continuous and categorical predictors. JSPI 89, 197-214.

  Koller, M. and Stahel, W. A. (2011). Sharpening Wald-type inference
  in robust regression for small samples. CSDA 55, 2504-2515.

  Koller, M. and Stahel, W. A. (2017). Nonsingular subsampling for
  regression S estimators with categorical predictors. Comp. Stat.
  32, 631-646.

Any divergences from `robustbase` are documented in
docs/numerical-notes.md.
