| 研究生: |
許晉瑋 Hsu, Chin-Wei |
|---|---|
| 論文名稱: |
以經驗分佈函數為基準之適合度檢定方法 Alternative Goodness-of-Fit Tests based on Empirical Distribution |
| 指導教授: |
洪英超
Hung, Ying-Chao |
| 口試委員: |
黃佳慧
Huang, Chia-Hui 曾能芳 Tseng, Neng-Fang |
| 學位類別: |
碩士
Master |
| 系所名稱: |
商學院 - 統計學系 Department of Statistics |
| 論文出版年: | 2020 |
| 畢業學年度: | 108 |
| 語文別: | 中文 |
| 論文頁數: | 46 |
| 中文關鍵詞: | 適合度檢定 、經驗分佈函數 、卡方適合度檢定 、Anderson-Darling檢定 、Kolmogorov-Smirnov檢定 |
| DOI URL: | http://doi.org/10.6814/NCCU202000071 |
| 相關次數: | 點閱:141 下載:16 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
適合度檢定為一種用以判斷某母體是否服從某特定分配的假設檢定,較為常用的一些適合度檢定有卡方適合度檢定,還有以經驗分佈函數(Empirical Distribution Function; EDF)為基準之適合度檢定,此類檢定的核心概念為評估經驗分佈函數與累積分佈函數(Cumulative Distribution Function; CDF)是否靠近,並以此建構合理的檢定統計量。此類檢定最為常用的為Anderson-Darling 檢定(A-D test)以及Kolmogorov-Smirnov 檢定(K-S test),A-D test 的檢定力普遍比K-S test 強,因其對分配的尾端較為敏感,但K-S test 執行起來較為簡單,亦可廣泛地延伸至多變量分配。本文主要是根據K-S test 的概念,定義一個稱為Lp-norm 的K-S 檢定統計量來執行連續型分配的適合度檢定。此方法可運用到單一變量及多變量分配的檢定,在電腦模擬的實驗下本文也證明所提方法於某些參數設定之下有較高的檢定力。
第一章 前言與文獻探討 1
第二章 檢定方法介紹 5
2.1 單變量適合度檢定 5
2.1.1 Kolmogorov-Smirnov檢定 5
2.1.2 Anderson-Darling檢定 8
2.2 多變量Kolmogorov-Smirnov檢定 10
2.3 以Lp-norm為基準之Kolmogorov-Smirnov檢定 14
2.3.1 針對單變量之Lp-norm K-S檢定 14
2.3.2 針對多變量之Lp-norm K-S檢定 16
第三章 模擬評估各檢定之表現 18
3.1 單變量適合度檢定 18
3.1.1 Location family的檢定 18
3.1.2 Scale family的檢定 26
3.2 雙變量適合度檢定 31
3.2.1 Location family的檢定 31
3.2.2 Scale family的檢定 37
第四章 結論 43
參考文獻 45
1. Alodat, M.T., Al-Subh, S.A., Ibrahim K., & Jemain A.A. (2010). “Empirical Characteristic Function Approach to Goodness of Fit Tests for the Logistic
Distribution under SRS and RSS”, Journal of Modern Applied Statistical Methods, Vol. 9, No. 2, 558-567.
2. Bakshaev, A., & Rudzkis, R. (2015). “Multivariate goodness-of-fit tests based on kernel density estimators”, Nonlinear Analysis: Modelling and Control, Vol. 20, No.4, 585-602.
3. Chen, W.C., Hung, Y.C., & Balakrishnan N. (2014) “Generating beta random numbers and Dirichlet random vectors in R: The package rBeta2009”, Computational Statistics and Data Analysis, 71, 1011-1020.
4. Facchinetti, S. (2009) “A Procedure to Find Exact Critical Values of Kolmogorov-Smirnov test”, Statistica Applicata – Italian Journal of Applied Statistics, Vol. 21, No. 3-4, 337-359.
5. Hung, Y.C., & Chen W.C. (2017). “Simulation of some multivariate distributions related to Dirichlet distribution with application to Monte Carlo simulations”,
Communication in Statistics-Simulation and Computation, Vol. 46, No. 6, 4281-4296.
6. Justel, A., Pena, D., & Zamar, R. (1997). “A Multivariate Kolmogorov-Smirnov Test of Goodness of Fit”, Statistics and Probability Letters, 35, 251-259.
7. McAssey, M.P. (2013) “An empirical goodness-of-fit test for multivariate distributions’, Journal of Applied Statistics, 40:5, 1120-1131.
8. Mirhossini, S.M., Amini M., & Dolati A. (2015) “On a general structure of bivariate FGM type distributions”, Application of Mathematics, Vol. 60, No. 1, 91-108.
9. Razali , N.M., & Wah, Y.B. (2011) “Power comparisons of Shapiro-Wilk,
Kolmogorov-Smirnov, Lillefors and Anderson-Darling tests”, Journal of Statistical Modeling and Analytics, Vol.2, No. 1, 21-33.
10. R package “Emcdf” (2018). URL: https://cran.r-project.org/web/packages/Emcdf/index.html.
11. R package “MultiRNG” (2019). URL: https://cran.r-project.org/web/packages/MultiRNG/index.html.
12. R package “pbivnorm” (2015). URL: https://github.com/brentonk/pbivnorm.
13. Stephens M.A. (1974). “EDF Statistics of Goodness of Fit and Some Comparisons”, Journal of the American Statistical Association, Vol. 69, No.347, 730-737.
14. Vaidyanathan, V.S., & Varghese, S. (2016) “Morgenstern type bivariate Lindley distribution”, Statistics, Optimization and Information Computing, Vol. 4, 132-146.
15. Yang, G.Y. (2012). “The Energy Goodness-of-Fit Test for Univariate Stable Distributions”.