Residual Analysis in the Nested and

碩士 === 國立高雄師範大學 === 數學系 === 96 === Abstract Residuals play an important role in the linear model. We can use residuals to detect model’s inadequacies and outliers. In the special cases of nested and split-plot designs, residuals form the identical equal pairs. Then we cannot use residuals to dis...

Full description

Bibliographic Details
Main Author: 李永娟
Other Authors: Pi-Hsiang Huang
Format: Others
Language:en_US
Published: 2008
Online Access:http://ndltd.ncl.edu.tw/handle/12869247995268879864
id ndltd-TW-096NKNU5479047
record_format oai_dc
spelling ndltd-TW-096NKNU54790472016-11-10T16:04:26Z http://ndltd.ncl.edu.tw/handle/12869247995268879864 Residual Analysis in the Nested and 巢式設計與裂區設計的殘差分析 李永娟 碩士 國立高雄師範大學 數學系 96 Abstract Residuals play an important role in the linear model. We can use residuals to detect model’s inadequacies and outliers. In the special cases of nested and split-plot designs, residuals form the identical equal pairs. Then we cannot use residuals to distinguish which point is an outlier from the model. In general, there are many measures to detect the outlier such as DFFITS, Cook’s distance, DFBETAS, Dixon’s test, Grubbs’ test, etc. Since these measures still form the equal pairs, we are unable to solve the problem with the existing methods. In this article, we discuss that the residuals form the identical equal pairs in the special cases of the nested and split-plot designs, and provide a way to solve this problem. Pi-Hsiang Huang 黃必祥 2008 學位論文 ; thesis 49 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立高雄師範大學 === 數學系 === 96 === Abstract Residuals play an important role in the linear model. We can use residuals to detect model’s inadequacies and outliers. In the special cases of nested and split-plot designs, residuals form the identical equal pairs. Then we cannot use residuals to distinguish which point is an outlier from the model. In general, there are many measures to detect the outlier such as DFFITS, Cook’s distance, DFBETAS, Dixon’s test, Grubbs’ test, etc. Since these measures still form the equal pairs, we are unable to solve the problem with the existing methods. In this article, we discuss that the residuals form the identical equal pairs in the special cases of the nested and split-plot designs, and provide a way to solve this problem.
author2 Pi-Hsiang Huang
author_facet Pi-Hsiang Huang
李永娟
author 李永娟
spellingShingle 李永娟
Residual Analysis in the Nested and
author_sort 李永娟
title Residual Analysis in the Nested and
title_short Residual Analysis in the Nested and
title_full Residual Analysis in the Nested and
title_fullStr Residual Analysis in the Nested and
title_full_unstemmed Residual Analysis in the Nested and
title_sort residual analysis in the nested and
publishDate 2008
url http://ndltd.ncl.edu.tw/handle/12869247995268879864
work_keys_str_mv AT lǐyǒngjuān residualanalysisinthenestedand
AT lǐyǒngjuān cháoshìshèjìyǔlièqūshèjìdecánchàfēnxī
_version_ 1718392086825271296