Care and Handling of Univariate Outliers in the General Linear Model to Detect Spuriosity—A Bayesian Approach
作者:
Irwin Guttman,
Rudolf Dutter,
PeterR. Freeman,
期刊:
Technometrics
(Taylor Available online 1978)
卷期:
Volume 20,
issue 2
页码: 187-193
ISSN:0040-1706
年代: 1978
DOI:10.1080/00401706.1978.10489645
出版商: Taylor & Francis Group
关键词: Outliers;Spurious observations;Posterior;Response surface designs;Multivariate-t density
数据来源: Taylor
摘要:
We deal with the situation covered by the univariate general linear model, that is, it is intended thatnobservations be generated in accordance with the usual model y =Xβ + ε however, it is feared thatkof the observations are spurious, that is, not generated in the manner intended, so that for an unknown set ofkdistinct integers, say (i1, …ik), a subset of the firstnintegers, we have, specifically, thatytj, =xtj′, β +aj, + ∊tj, where in generalxt′ denotes thet-th row ofX, and where (a1, …,ak), so called shift parameters, are such that – ∞ <aj, < ∞. In this paper, we discuss the posterior distribution of β, when indeed it is assumed a priori that any given set ofkobservations has uniform probability I/(nk) of being spurious. The properties of the posterior of β are discussed, and the results used in an example using data generated from a response surface design. Ad hoc procedures are discussed for gaining information onk, whenkis unknown. These ad hoc procedures are illustrated using a famous set of data from Darwin
点击下载:
PDF (559KB)
返 回