A Detection Method of Multivariate Outliers using Decompositions of the Squared Mahalanobis Distance

• 발행기관: 한국자료분석학회
• 저자명: 정강모(Kang-Mo Jung)
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.7 No.6, 1935~1943쪽, 전체 9쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2005.12.30

영문 초록

Kim(2000) proposed two meaningful decompositions of the squared Mahalanobis distance to uncover the sources of outlyingness for multivariate observations. The decompo- sition is useful for identifying some component variables dominating the Mahalanobis distance. In this article we considered the distributions for components of the decompo- sitions and we showed that each component follows a normal distribution and the random vector consisting of two components is a multivariate normal distribution. We may detect outliers using the cut-off values. We proposed a graphical tool for detecting multivariate outliers through one-sheet figure. It is a very useful result to analyze the outlyingness for high dimensional data. Two illustrative examples are given.

목차

1. Introduction

2. Distributions for components of decompositions of the squared Mahalanobis distance

3. Graphical Results

4. Numerical Examples

5. Concluding Remarks

References

키워드

cut-off value decomposition Mahalanobis distance multivariate outlier