A Fast Algorithm for the Minimum Covariance Determinant Estimator

By Peter J. Rousseeuw, Katrien van Driessen. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Sever: A Robust Meta-Algorithm for Stochastic Optimization (2018, methodology): “The minimum covariance determinant estimator [RD99] only applies when the number of data points n exceeds 2d, which does not hold for the datasets we consider (it also has other issues such as computational intractability).”
Image-Based Phenotyping of the Mature Arabidopsis Shoot System (2014, methodology): “If the number of leaves on a plant is greater or equal 20, the robust Minimum Covariance Determinant estimator is used (outlier detection based on all three features) [25].”
Inconsistency of Resampling Algorithms for High-Breakdown Regression Estimators and a New Algorithm (2002, methodology): “A simplified version of the LTS(c) algorithms of Ruppert (1992), Hawkins and Olive (1999) and Rousseeuw and Van Driessen (1999b) uses K elemental starts.”
ROBPCA: A New Approach to Robust Principal Component Analysis (2005, methodology): “…matrix has minimal determinant Because in general we cannot consider all h -subsets, we must rely on approximate algorithms Here we slightly adapt the FAST–MCD algorithm of Rousseeuw and Van Driessen (1999) by taking advantage of the result of stage 2, which used the outlyingness measure (A.3) 1.”
Anomaly Detection in Time Series: A Comprehensive Evaluation (2022, methodology): “reconstruction original Pytorch Sub-Fast-MCD [115] Statistics distribution own Python”
Equivariance and invariance properties of multivariate quantile and related functions, and the role of standardisation (2010, background): “sense that T (ax + b, FaX + b) = T (x, FX) (37) for any scalar a and any vector b.”
Outlier identification in high dimensions (2008, background): “ Rousseeuw and van Driessen (1999) developed a faster version of MCD which was a sizable improvement, but is still quite computationally intensive.”
A robust Hotelling test (2002, methodology): “Earlier analysis of this data ([10]) showed a strongly deviating group of outliers, ranging from index 491 to index 565.”
A State of the Art Review of Modal-Based Damage Detection in Bridges: Development, Challenges, and Solutions (2017, methodology): “A further advancement to the standard MSD is to pair it with the Minimum Covariance Determinate (MCD) estimator [99] that mitigates the uncertainty regarding the presence of outliers in training data.”
The Distribution of Robust Distances (2005, background): “Calculating the MCD can be quite computationally intensive.”