2005
DOI: 10.1198/106186005x77685
|View full text |Cite
|
Sign up to set email alerts
|

The Distribution of Robust Distances

Abstract: Mahalanobis-type distances in which the shape matrix is derived from a consistent, high-breakdown robust multivariate location and scale estimator have an asymptotic chisquared distribution as is the case with those derived from the ordinary covariance matrix. For example, Rousseeuw's minimum covariance determinant (MCD) is a robust estimator with a high breakdown. However, even in quite large samples, the chi-squared approximation to the distances of the sample data from the MCD center with respect to the MCD… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
136
0
3

Year Published

2008
2008
2021
2021

Publication Types

Select...
6
2

Relationship

0
8

Authors

Journals

citations
Cited by 173 publications
(142 citation statements)
references
References 35 publications
3
136
0
3
Order By: Relevance
“…Hence, a point of Mahalanobis squared distance of 3 or less lies within the boundary of 99% of data. It is well documented that the MD is described by a chi-square (χ 2 ) distribution with a degree of freedom (df) equivalent to the number of independent variables (Hardin and Rocke [36]). A tested weather generator is called a goodfit candidate when the reference point of observations (represented by its features) falls within a reasonable distance from the investigated data center.…”
Section: Mahalanobis Distancementioning
confidence: 99%
“…Hence, a point of Mahalanobis squared distance of 3 or less lies within the boundary of 99% of data. It is well documented that the MD is described by a chi-square (χ 2 ) distribution with a degree of freedom (df) equivalent to the number of independent variables (Hardin and Rocke [36]). A tested weather generator is called a goodfit candidate when the reference point of observations (represented by its features) falls within a reasonable distance from the investigated data center.…”
Section: Mahalanobis Distancementioning
confidence: 99%
“…MCD algoritmasının çalışması için gereken zaman değişken ve gözlem sayısına bağlı olarak oldukça fazla olabilir. Pek çok araştırmacı bu problemin üstesinden gelebilmek için algoritmayı iyileştirmeye/hızlandırmaya odaklanmışlardır [32][33][34][35]. Son olarak Rousseeuw ve Van Driessen [34] tarafından hazırlanan hızlı bir algoritma ile son derece cazip hale gelen sağlam faktör analizi aşağıdaki temel özelliklere sahip olmuştur:…”
Section: Sağlam Faktör Analizi (Robust Factor Analysis)unclassified
“…h değerinin aykırı değer içermeyen minimum gözlem sayısından oluştuğu kabul edilir. Seçilecek h adet gözlem için hesaplanan ortalama MCD'nin konum parametre kestirimi, aynı gözlemler için hesaplanan varyanskovaryans matrisi de ölçek parametre kestirimi olacaktır [23], [35], [37].…”
Section: Minimum Kovaryans Determinantı Metodu (Minimum Covariance Deunclassified
See 1 more Smart Citation
“…Khattree & Naik (1995) used a Q-Q plot based on a modification of the Mahalanobis distance. Hadi (1992) and Hardin & Rocke (2005) used a robustified version of the Mahalanobis distance for detecting multivariate outliers. Muruzábal & noz (1997) proposed a graphical technique called SelfOrganizing Maps as a tool for detecting multivariate plots.…”
Section: Introductionmentioning
confidence: 99%