The Distribution of Robust Distances

By Johanna S. Hardin, David M. Rocke. 2005

216 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/1910eb29c724c900ddb9dafac435e0ac23bc0a05

How researchers see this paper

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

In Statistics and Probability

Citations in the corpus

Rainbow Plots, Bagplots, and Boxplots for Functional Data (2010, background): “See Becker & Gather (2001) and Hardin & Rocke (2005) for discussion on variations of this method.”
MacroBase: Prioritizing Attention in Fast Data (2016, methodology): “We view existing explanation techniques as a useful second step in analysis following the explanations generated by MacroBase, and we see promise in adapting these existing techniques to streaming execution at high volume.”
Outlier detection for high-dimensional data (2015, unknown): “To determine the cut-off value for outlying points, Hardin & Rocke (2005) provided a distributional result for the Mahalanobis distance in (1) based on the minimum covariance determinant estimator.”
Outlier identification in high dimensions (2008, unknown): “Also, if robust estimators T and C are applied to a large data set in which the non-outliers are normally distributed, Hardin and Rocke (2005) found that the squared distances could be described by a scaled F -distribution.”
Finding an unknown number of multivariate outliers (2009, methodology): “MCD/i; scaled F -distribution of Hardin and Rocke (2005) PP Squared projection-based robust distances of Peña and Prieto (2001); scaled F -distribution”
MacroBase: Analytic Monitoring for the Internet of Things (2016, background): “s in the body of the distribution—sufﬁces to identify many important classes of outliers in the applications we study. The robust statistics literature discusses this phenomena in the case of the MCD [41] and echoes Aggarwal’s observation that “even for arbitrary distributions, [extreme value analys”
Outlier Detection for Compositional Data Using Robust Methods (2008, background): “It is common to use the same cut-off value from the χ2d distribution (Rousseeuw and Van Zomeren, 1990), although other approximations could lead to more accurate cut-off values (Filzmoser, Garrett, and Reimann, 2005; Hardin and Rocke, 2005).”
A Multivariate Robust Control Chart for Individual Observations (2009, methodology): “In this paper, we proposed a multivariate robust Hotelling T 2 chart based on reweighted MCD estimates as an alternative to the classical multivariate T 2 control charts for Phase II data. The proposed control chart is obtained by replacing the classical mean vector and covariance matrix of the data”