The Convergence of Efroymson's Stepwise Regression Algorithm
作者:
AlanJ. Miller,
期刊:
The American Statistician
(Taylor Available online 1996)
卷期:
Volume 50,
issue 2
页码: 180-181
ISSN:0003-1305
年代: 1996
DOI:10.1080/00031305.1996.10474372
出版商: Taylor & Francis Group
关键词: Stepwise regression;Stopping rule
数据来源: Taylor
摘要:
The stepwise regression algorithm that is widely used is due to Efroymson. He stated that theF-to-remove value had to be not greater than theF-to-enter value, but did not show that the algorithm could not cycle. Until now nobody appears to have shown this. To prove that the algorithm does converge, an objective function is introduced. It is shown that this objective function decreases or can occasionally remain constant at each step in the algorithm, and hence the algorithm cannot cycle provided that Efroymson's condition is satisfied.
点击下载:
PDF (189KB)
返 回