Benchmark testing of algorithms for very robust regression: FS, LMS and LTS

详细信息	查看全文 | 推荐本文 |

• 作者：Francesca Torti^a ; ^{francesca.torti@nemo.unipr.it} ; Domenico Perrotta^b ; ^{domenico.perrotta@ec.europa.eu} ; Anthony C. Atkinson^c ; ^{a.c.atkinson@lse.ac.uk} ; Marco Riani^d ; ^{mriani@unipr.it}
• 关键词：Combinatorial search ; Concentration step ; Forward search ; Least median of squares ; Least trimmed squares ; Logistic plots of power ; Masking ; Outlier detection
• 刊名：Computational Statistics and Data Analysis
• 出版年：2012
• 期刊代码：19_01679473
• 类别：cp
• 出版时间：August, 2012
• 卷：56
• 期：8
• 页码：2501-2512
• 文件大小：373 K

摘要

The methods of very robust regression resist up to 50%of outliers. The algorithms for very robust regression rely on selecting numerous subsamples of the data. New algorithms for LMS and LTS estimators that have increased computational efficiency due to improved combinatorial sampling are proposed. These and other publicly available algorithms are compared for outlier detection. Timings and estimator quality are also considered. An algorithm using the forward search (FS) has the best properties for both size and power of the outlier tests.