A note on lack-of-fit tests for linear models without replication

• 成果类型: Article
• 署名作者: Su, ZH; Yang, SS
• 署名单位: Harvard University; Harvard T.H. Chan School of Public Health; Kansas State University
• 刊物名称: JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
• ISSN/ISSBN: 0162-1459
• DOI: 10.1198/016214505000000709
• 发表日期: 2006
• 页码: 205-210
• 关键词: maximin clusters; regression

摘要

A class of three tests-overall lack-of-fit test, between-cluster lack-of-fit test. and within-cluster lack-of-fit test-are proposed for testing the lack of fit of a linear regression model applied to experiments without replicates. The power of the proposed tests is significantly higher than those of the known tests under the situations considered here. The proposed tests are capable of detecting which type of lack of fit is dominant when both between-cluster and within-cluster lack of fit are present.