Finding an unknown number of multivariate outliers

• 成果类型: Article
• 署名作者: Riani, Marco; Atkinson, Anthony C.; Cerioli, Andrea
• 署名单位: University of London; London School Economics & Political Science; University of Parma
• 刊物名称: JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
• ISSN/ISSBN: 1369-7412
• DOI: 10.1111/j.1467-9868.2008.00692.x
• 发表日期: 2009
• 页码: 447-466
• 关键词: multiple outliers; covariance; identification; asymptotics; matrix; estimators; algorithm; points

摘要

We use the forward search to provide robust Mahalanobis distances to detect the presence of outliers in a sample of multivariate normal data. Theoretical results on order statistics and on estimation in truncated samples provide the distribution of our test statistic. We also introduce several new robust distances with associated distributional results. Comparisons of our procedure with tests using other robust Mahalanobis distances show the good size and high power of our procedure. We also provide a unification of results on correction factors for estimation from truncated samples.