1. NCCU Academic Hub
  2. Publications
  3. College of Commerce
  4. Theses
  5. Item

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 The Box-Cox 依變數轉換之技巧
The Box-Cox Transformation: A Review
作者 曾能芳
Chan, Lan Fun
貢獻者 鄭天澤
曾能芳
Chan, Lan Fun
關鍵詞 強力轉換
常態化轉換
架構變數
分數檢定
敏感性
power transformation
normalized transformation
constructed variable
scores test
Andrews`s exact test
sensitivity
日期 1996
上傳時間 28-Apr-2016 11:48:22 (UTC+8)
摘要 　　The use of transformation can usually simplify the analysis of data, especiallywhen the original observations deviate from the underlying assumption of linearmodel. Box-Cox transformation receives much more attention than others. Inthis dissertation, we will review the theory about the estimation, hypotheses test on transformation parameter and about the sensitivity of the linear model parameters. Monte Carlo simulation is used to study the performance of the transformation. We also display whether Box-Cox transformation makes the transformed observations satisfy the assumption of linear model actually.
描述 碩士
國立政治大學
統計學系
82354003
資料來源 http://thesis.lib.nccu.edu.tw/record/#B2002002785
資料類型 thesis
dc.contributor.advisor 鄭天澤zh_TW
dc.contributor.author (Authors) 曾能芳zh_TW
dc.contributor.author (Authors) Chan, Lan Funen_US
dc.creator (作者) 曾能芳zh_TW
dc.creator (作者) Chan, Lan Funen_US
dc.date (日期) 1996en_US
dc.date.accessioned 28-Apr-2016 11:48:22 (UTC+8)-
dc.date.available 28-Apr-2016 11:48:22 (UTC+8)-
dc.date.issued (上傳時間) 28-Apr-2016 11:48:22 (UTC+8)-
dc.identifier (Other Identifiers) B2002002785en_US
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/87300-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 統計學系zh_TW
dc.description (描述) 82354003zh_TW
dc.description.abstract (摘要) 　　The use of transformation can usually simplify the analysis of data, especiallywhen the original observations deviate from the underlying assumption of linearmodel. Box-Cox transformation receives much more attention than others. Inthis dissertation, we will review the theory about the estimation, hypotheses test on transformation parameter and about the sensitivity of the linear model parameters. Monte Carlo simulation is used to study the performance of the transformation. We also display whether Box-Cox transformation makes the transformed observations satisfy the assumption of linear model actually.en_US
dc.description.tableofcontents 謝辭
Abstract
CONTENTS
Chapter 1　INTRODUCTION-----1
Chapter 2　POWER TRANSFORMATION-----3
　　2.1　Estimation of λ-----3
　　　　2.1.1　Box-Cox method-----3
　　　　2.1.2　Atkinson`s method-----4
　　2.2　Asymptotic properties of λ, θ(λ) and σ(λ)-----6
　　2.3　Hypotheses tests about λ-----6
　　　　2.3.1　Likelihood ratio test-----7t
　　　　2.3.2　The scores test-----7
　　　　2.3.3　Andrews`s test-----8
　　2.4　Randomness of λ-----9
　　　　2.4.1　λ regarded as random-----10
　　　　2.4.2　λ regarded as fixed-----12
Chapter 3　SENSITIVITY FOR LINEAR MODEL PARAMETER-----14
　　3.1　Stability for contrast parameter-----15
　　3.2　Sensitivity for location parameter-----17
　　3.3　Simulation-----19
Chapter 4　TRANSFORMATION TO SYMMETRY-----25
　　4.1　Skewness of yi(λ) with non-normal distribution-----25
　　4.2　Precision of the normal-theory estimate λN-----26
　　4.3　Estimate of new λ to symmetry-----27
Chapter 5　CONCLUSION-----30
APPENDIX A　Asymptotic maximum likelihood theory-----32
APPENDIX B　Normal-theory maximum likelihood-----34
REFERENCE-----35		zh_TW
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#B2002002785en_US
dc.subject (關鍵詞) 強力轉換zh_TW
dc.subject (關鍵詞) 常態化轉換zh_TW
dc.subject (關鍵詞) 架構變數zh_TW
dc.subject (關鍵詞) 分數檢定zh_TW
dc.subject (關鍵詞) 敏感性zh_TW
dc.subject (關鍵詞) power transformationen_US
dc.subject (關鍵詞) normalized transformationen_US
dc.subject (關鍵詞) constructed variableen_US
dc.subject (關鍵詞) scores testen_US
dc.subject (關鍵詞) Andrews`s exact testen_US
dc.subject (關鍵詞) sensitivityen_US
dc.title (題名) The Box-Cox 依變數轉換之技巧zh_TW
dc.title (題名) The Box-Cox Transformation: A Reviewen_US
dc.type (資料類型) thesisen_US