RDELA—a Delaunay-triangulation-based, location and covariance estimator with high breakdown point

被引:0
作者
Steffen Liebscher
Thomas Kirschstein
Claudia Becker
机构
[1] Martin-Luther-University,
来源
Statistics and Computing | 2013年 / 23卷
关键词
Breakdown point; Delaunay triangulation; Minimum covariance determinant; Robust estimation;
D O I
暂无
中图分类号
学科分类号
摘要
We propose an approach that utilizes the Delaunay triangulation to identify a robust/outlier-free subsample. Given that the data structure of the non-outlying points is convex (e.g. of elliptical shape), this subsample can then be used to give a robust estimation of location and scatter (by applying the classical mean and covariance). The estimators derived from our approach are shown to have a high breakdown point. In addition, we provide a diagnostic plot to expand the initial subset in a data-driven way, further increasing the estimators’ efficiency.
引用
收藏
页码:677 / 688
页数:11
相关论文
共 53 条
[41]  
Willems G.(undefined)undefined undefined undefined undefined-undefined
[42]  
Riani M.(undefined)undefined undefined undefined undefined-undefined
[43]  
Atkinson A.(undefined)undefined undefined undefined undefined-undefined
[44]  
Cerioli A.(undefined)undefined undefined undefined undefined-undefined
[45]  
Rocke D.(undefined)undefined undefined undefined undefined-undefined
[46]  
Rocke D.(undefined)undefined undefined undefined undefined-undefined
[47]  
Woodruff D.(undefined)undefined undefined undefined undefined-undefined
[48]  
Rousseeuw P.(undefined)undefined undefined undefined undefined-undefined
[49]  
Rousseeuw P.(undefined)undefined undefined undefined undefined-undefined
[50]  
van Driessen K.(undefined)undefined undefined undefined undefined-undefined