GitHub

LinRegOutliers

A Julia package for outlier detection in linear regression.

Implemented Methods

  • Ordinary Least Squares and Weighted Least Squares regression
  • Regression diagnostics (DFBETA, DFFIT, CovRatio, Cook's Distance, Mahalanobis, Hadi's measure, etc.)
  • Hadi & Simonoff (1993)
  • Kianifard & Swallow (1989)
  • Sebert & Montgomery & Rollier (1998)
  • Least Median of Squares
  • Least Trimmed Squares
  • Minimum Volume Ellipsoid (MVE)
  • MVE & LTS Plot
  • Billor & Chatterjee & Hadi (2006)
  • Pena & Yohai (1995)
  • Satman (2013)
  • Satman (2015)
  • Setan & Halim & Mohd (2000)
  • Least Absolute Deviations (LAD)
  • Quantile Regression Parameter Estimation (quantileregression)
  • Least Trimmed Absolute Deviations (LTA)
  • Hadi (1992)
  • Marchette & Solka (2003) Data Images
  • Satman's GA based LTS estimation (2012)
  • Fischler & Bolles (1981) RANSAC Algorithm
  • Minimum Covariance Determinant Estimator
  • Imon (2005) Algorithm
  • Barratt & Angeris & Boyd (2020) CCF algorithm
  • Atkinson (1994) Forward Search Algorithm
  • BACON Algorithm (Billor & Hadi & Velleman (2000))
  • Hadi (1994) Algorithm
  • Chatterjee & Mächler (1997)
  • Theil-Sen estimator for multiple regression
  • Deepest Regression Estimator
  • Robust Hat Matrix based Initial Subset Regressor
  • Summary

Algorithms

Installation

LinRegOutliers can be installed using the Julia REPL. 

julia> ]
(@v1.8) pkg> add LinRegOutliers

or

julia> using Pgk
julia> Pkg.add("LinRegOutliers")

then

julia> using LinRegOutliers

to make all the stuff be ready!

Examples

We provide some examples here.

Documentation

Please check out the reference manual here.

News

  • We implemented algorithm(X, y) style calls for all of the algorithms where X is the design matrix and y is the response vector.
  • We implemented ~25 outlier detection algorithms which covers a high percentage of the literature.

Contributions

You are probably the right contributor

  • If you have statistics background
  • If you like Julia

However, the second condition is more important because an outlier detection algorithm is just an algorithm. Reading the implemented methods is enough to implement new ones. Please follow the issues. Here is the a bunch of first shot introductions for new comers. Welcome and thank you in advance!

Citation

Please refer our original paper if you use the package in your research using

Satman et al., (2021). LinRegOutliers: A Julia package for detecting outliers in linear regression. Journal of Open Source Software, 6(57), 2892, https://doi.org/10.21105/joss.02892

or the bibtex entry

@article{Satman2021,
  doi = {10.21105/joss.02892},
  url = {https://doi.org/10.21105/joss.02892},
  year = {2021},
  publisher = {The Open Journal},
  volume = {6},
  number = {57},
  pages = {2892},
  author = {Mehmet Hakan Satman and Shreesh Adiga and Guillermo Angeris and Emre Akadal},
  title = {LinRegOutliers: A Julia package for detecting outliers in linear regression},
  journal = {Journal of Open Source Software}
}

Contact & Communication

  • Please use issues for a new feature request or bug reports.
  • We are in #linregoutliers channel on Julia Slack for any discussion requires online chatting.