跳到主要內容

簡易檢索 / 詳目顯示

研究生: 陳彬
Ben Chen
論文名稱: 型1設限下韋伯參數估計問題
指導教授: 呂理裕
Lii-Yuh Leu
口試委員:
學位類別: 碩士
Master
系所名稱: 理學院 - 統計研究所
Graduate Institute of Statistics
畢業學年度: 91
語文別: 中文
論文頁數: 53
中文關鍵詞: 韋伯分配極值分配信賴區間型1設限
相關次數: 點閱:20下載:0
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
  • 關於型I設限資料下,韋伯分配參數的估計方法,本文採用最大概似估計量作為參數的點估計,而區間估計方面,除了傳統上利用大樣本性質,以近似常態法估計信賴區間外,另考量分別以log-likelihood檢定統計量為基礎、以及Bootstrap方法為基礎,共呈現十種方法估計參數之信賴區間。
    為比較各種方法之優劣,本文透過模擬範例,以區間之平均長度和覆蓋機率作為比較的依據。結果顯示當樣本數為小樣本時,傳統上採用近似常態法並非最好的估計方式,反觀以log-likelihood檢定統計量為基礎的四種方法,即使在觀察之樣本數少的時候,覆蓋機率皆可達到名義之水準,當中的SLR和Third-Order方法更可得到較小的信賴區間,因此是較佳的估計方式。在以Bootstrap為基礎的五種方法中,PBSRLLR方法表現良好,當觀察之樣本數少的時候,覆蓋機率可達到名義之水準,其他四種Bootstrap方法則不能。在區間長度的表現上,當觀察之樣本數小於9個的時候,PBSRLLR可以得到比SLR和Third-Order方法更小的信賴區間,在此條件下,PBSRLLR是較佳的估計方式。


    第一章 緒論………………………………………………………………….1  1-1 研究動機與文獻回顧…………………………………………………1  1-2 本文架構………………………………………………………………2 第二章 韋伯分配參數的最大概似估計量………………………………….4  2-1 透過韋伯分配計算最大概似估計量…………………………………4  2-2 透過極值分配計算最大概似估計量…………………………………6 第三章 參數信賴區間的估計方法………………………………………….9  3-1 NORM:Normal- Approximation Procedures………………………..10  3-2 LLR:Log-likelihood Ratio Procedures………………………………13  3-3 SLR:A Mean and Variance Correction to the Signed Square Roots       LLR Procedures……………………………………………….15  3-4 BLR:Bartlett Correction to the LLR Procedures……………………17  3-5 Third-Order:A likelihood Based Third-Order Procedures………….18  3-6 PBP:Parametric Bootstrap Percentile Procedures…………………..22  3-7 PBBC:Parametric Bootstrap Bias-correct Procedures……………...24  3-8 PBBCA:Parametric Bootstrap Bias-correct Acceleration Procedure.26  3-9 PBT:Parametric Bootstrap-t Procedures…………………………….29  3-10 PBSRLLR:Parametric Bootstrap Sign Square Root LLR ………...32 第四章 模擬範例…………………………………………………………...35  4-1 模擬實驗的架構……………………………………………………..35  4-2 模擬試驗的結果……………………………………………………..38   4-2-1 形狀參數β的模擬結果…………………………………………38   4-2-2 百分位點的模擬結果…………………………………………...42 第五章 結論………………………………………………………………...49 參考文獻……………………………………………………………………...52 附錄1 SLR中m和s的算法………………………………………………53

    Bain, L. J. (1978), Statistical Analysis of Reliability and Life-Testing Models,
      New York: Marcel Dekker, Inc..
    Barndorff-Nielsen, O. E. (1986), “Inference on Full or Partial Parameters
      Based on the Standardized Signed Log Likelihood Ratio,” Bimetrika, 73,
      307-322.
    DiCiccio, T. J. (1987), "Approximate Inference for the Generalized Gamma
      Distribution," Technometrics, 29, 33-40.
    DiCiccio, T. J. (1988), "Likelihood Inference for Linear Regression Models,"
      Biometrika, 75, 29-34.
    DiCiccio, T. J., and Martin, M. A. (1990), "Approximations of Marginal Tail
      Probabilities for a Class of Smooth Functions With Applications to
      Bayesian and Conditional Inference," Biometrika, 78, 891-902.
    Doganaksoy, N., and Schmee, J. (1993), “Comparisons of Approximate
      Confidence Intervals for Distributions Used in Life-data Analysis,”
      Technometrics, 35, 175-184.
    Efron, B., and Tibshirani, R. J. (1993), An Introduction to the Bootstrap, New
      York: Chapman & Hall.
    Jeng, S.L., and Meeker, W. Q. (2000), “Comparisons of Approximate
      Confidence Interval Procedures for Type I Censored Data,” Technometrics
      ,42, 135-148.
    Lawless, J. F. (1993),Statistical Methods for Reliability Data, New York: Wiley.
    Monen, M. V. (1963), “Estimation of the Shape and Scale Parameters of the
      Weibull Distribution,” Technometrics, 5, 175-185.
    Nelson, W. (1982), Applied Life Data Analysis. New York: Wiley.
    Ostrouchov, G.., and Meeker, W. Q, Jr. (1988), “Accuracy of Approximate
      Confidence Bounds Computed From Interval Censored Weibull or
      Lognormal Data,” Journal of Statistical Computation and Simulation,
      29,43-76.
    Wong, A., and Wu, J. (2000), “Practical Small-Sample Asymptotics for
      Distributions Used in Life-Data Analysis,” Technometrics, 42, 149-155.

    QR CODE
    :::