透過您的圖書館登入
IP:18.119.133.96
  • 學位論文

區間設限資料下對分佈函數之研究

Distribution Function Estimation for Interval-Censored Data

指導教授 : 吳裕振

摘要


本論文欲探討兩個檢查時間點,即型二區間設限資料下,對存活累積分佈函數之研究分析。其中使用伯氏多項式來描述其累積分佈函數,進而建構出概似函數。 我們使用最大概似方法來估計其參數,由於求取其參數之最大概似估計並不容易,故使用馬可夫鏈蒙地卡羅法─Simulated Annealing來估計,其模擬結果,相當不錯。

並列摘要


We study two check-out times in this paper. In other words, we using Case Ⅱ interval-censored data analysis of the survival cumulative distribution function. In the course of research we refer to Chang et al. (2005) that Bernstein Polynomial to describe the cumulative distribution function. Further, we find out likelihood function. We use the method of maximum likelihood estimation (MLE) is not an easy task. Therefore, we will use Markov chain Monte Carlo - Simulated Annealing to estimate the parameters, and the simulate results are quite good.

參考文獻


[6 ]楊雅雯 (2009) 使用伯氏多項式對存活現狀數據之最大概似估計,
[7 ]賴京鈺 (2011) 存活現狀數據下風險函數之最大概似估計,中原大
[8 ]許時淮 (2015) 現狀數據資料下勝算比之研究,中原大學碩士論文
[2 ]P. Groeneboom (1991), “Nonparametric maximum likelihood estimators for interval censoring and deconvolution ”, Technical Report, Department of Statistics, Standford University
[3 ]P.J. Green (1995), “ Reversible jump Markov chain Monte Carlo computation and Bayesian model determination ”, Biometrika, Vol. 82, P.711-732

延伸閱讀