REDUCED DISTANCE MEASURES AND TRANSFORMATIONS IN PROCESSING MULTIVARIATE OUTLIERS
作者:
Vic Barnett,
期刊:
Australian Journal of Statistics
(WILEY Available online 1983)
卷期:
Volume 25,
issue 1
页码: 64-75
ISSN:0004-9581
年代: 1983
DOI:10.1111/j.1467-842X.1983.tb01197.x
出版商: Blackwell Publishing Ltd
关键词: Multivariate outliers;Tests of discordancy;Normal data;Non‐normal data;Reduced ordering;Reduced measures transformations;Bivariate exponential distribution;Bivariate Pareto distribution
数据来源: WILEY
摘要:
SummaryThe study of multivariate outliers raises many problems of definition, principle and manipulation. Well‐authenticated tests of discordancy exist only for the multivariate normal distribution. Detection of outliers in non‐normal distributions involves the adoption of appropriate criteria to represent ‘extremeness' of observations in a sample; corresponding tests of discordancy usually require tedious, or even intractable, distributional and computational manipulations. A class of transformations of the data is considered with a view of transferring some of the familiar and desirable features of discordancy tests for normal samples to non‐normal sit
点击下载:
PDF
(524KB)
返 回