A Comparative Study of Regression Concerning Weighted Least Squares Methods
作者:
Costel Sǎrbu,
期刊:
Analytical Letters
(Taylor Available online 1995)
卷期:
Volume 28,
issue 11
页码: 2077-2094
ISSN:0003-2719
年代: 1995
DOI:10.1080/00032719508000026
出版商: Taylor & Francis Group
关键词: Calibration;robust and weighted regression
数据来源: Taylor
摘要:
The weighted least squares method using 1/x2as a weighting factor is described and compared with conventional ordinary and weighted least squares and robust regression. Applications of these different methods to the relevant data sets demonstrates that the performance of the procedure discussed in this paper exceeds that of ordinary least squares method and equals, and often exceeds, that of weighted or robust methods.
点击下载:
PDF (534KB)
返 回