LIBRA: a MATLAB Library for Robust Analysis is developed at ROBUST@Leuven, the research group on robust statistics at the KU Leuven. It contains user-friendly implementations of several robust procedures. These methods are resistant to outliers in the data. Currently, the library contains functions for univariate location, scale and skewness, multivariate location and covariance estimation (MCD), regression (LTS, MCD-regression), Principal Component Analysis (RAPCA, ROBPCA), Principal Component Regression (RPCR), Partial Least Squares Regression (RSIMPLS), classification (RDA, RSIMCA), clustering, outlier detection for skewed data (including the bagplot based on halfspace depth), censored depth quantiles and detecting cellwise outliers (DDC). For comparison also several non-robust functions are included. Many graphical tools are provided for model checking and outlier detection.
Most of the functions require the MATLAB Statistics Toolbox.
Contributions to this library have been made by (in alphabetical order): Guy Brys, Michiel Debruyne, Sanne Engelen, Mia Hubert, Wai Yan Kong, Nele Smets, Wannes Van den Bossche, Karlien Vanden Branden, Stephan Van der Veeken, Ellen Vandervieren, Katrien Van Driessen, Sabine Verboven, Tim Verdonck en Fabienne Verwerft.
The library can be freely used for non-commercial use only. Please make appropriate references to the corresponding paper(s) if you use any of our programs. The correct references can be found in the help-files, or at the webpage:
More details on the use of the library are described in:
- Verboven, S., Hubert, M. (2005), LIBRA: a MATLAB Library for Robust Analysis, 'Chemometrics and Intelligent Laboratory Systems', 75, 127-136. [pdf]
- Verboven, S., Hubert, M. (2010). Matlab library LIBRA, 'Wiley Interdisciplinary Reviews: Computational Statistics', 2, 509-515. [pdf]
Bugs or comments on the programs can be reported to Mia Hubert.
Several functions of the LIBRA Toolbox are also available in the PLS Toolbox at Eigenvector Research.
The current list of functions can be found here.
The new release of June 28, 2016 now contains DDC.m (detecting cellwise cells).
Download LIBRA for
Last update of LIBRA: June 28, 2016.