Outlier detection in non-elliptical data by kernel MRCD
From MaRDI portal
Abstract: The minimum regularized covariance determinant method (MRCD) is a robust estimator for multivariate location and scatter, which detects outliers by fitting a robust covariance matrix to the data. Its regularization ensures that the covariance matrix is well-conditioned in any dimension. The MRCD assumes that the non-outlying observations are roughly elliptically distributed, but many datasets are not of that form. Moreover, the computation time of MRCD increases substantially when the number of variables goes up, and nowadays datasets with many variables are common. The proposed Kernel Minimum Regularized Covariance Determinant (KMRCD) estimator addresses both issues. It is not restricted to elliptical data because it implicitly computes the MRCD estimates in a kernel induced feature space. A fast algorithm is constructed that starts from kernel-based initial estimates and exploits the kernel trick to speed up the subsequent computations. Based on the KMRCD estimates, a rule is proposed to flag outliers. The KMRCD algorithm performs well in simulations, and is illustrated on real-life data.
Recommendations
- The minimum regularized covariance determinant estimator
- Kernel density outlier detector
- The feasible solution algorithm for the minimum covariance determinant estimator in multivariate data
- Modified minimum covariance determinant estimator and its application to outlier detection of chemical process data
- Multivariate outlier detection with high-breakdown estimators
Cites work
- A kernel two-sample test
- A Measure of Directional Outlyingness With Applications to Image Data and Video
- Alternatives to the Median Absolute Deviation
- An introduction to copulas.
- Detecting influential observations in kernel PCA
- High-breakdown robust multivariate methods
- scientific article; zbMATH DE number 3986407 (Why is no real title available?)
- scientific article; zbMATH DE number 194744 (Why is no real title available?)
- scientific article; zbMATH DE number 1843268 (Why is no real title available?)
- Influence function and efficiency of the minimum covariance determinant scatter matrix estimator
- Kernel ellipsoidal trimming
- Least Median of Squares Regression
- Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination
- Robust kernel principal component analysis and classification
- Sign and rank covariance matrices
- The minimum regularized covariance determinant estimator
- The multivariate L 1 -median and associated data depth
Cited in
(13)- Kernel ellipsoidal trimming
- Identifying outliers using multiple kernel canonical correlation analysis with application to imaging genetics
- Kernel density outlier detector
- Consistency factor for the MCD estimator at the Student-\(t\) distribution
- Fast Robust Location and Scatter Estimation: A Depth-based Method
- The Cellwise Minimum Covariance Determinant Estimator
- Robust distances and multivariate outlier detection under heavy tails
- An adaptive threshold for outlier detection in high-dimensional settings
- CellMCD+: an improved outlier-resistant cellwise minimum covariance determinant method
- A Stability Framework for Parameter Selection in the Minimum Covariance Determinant Problem
- Spatial depth for data in metric spaces
- Weighted embedding and outlier detection of metric space data
- The minimum regularized covariance determinant estimator
This page was built for publication: Outlier detection in non-elliptical data by kernel MRCD
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2058886)
