透過您的圖書館登入
IP:3.145.97.248
  • 學位論文

使用伯氏多項式對現狀數據勝算比模型之最大概似估計

Maximum Likelihood Estimator of Proportional Odds Model with Current Status Data Using Bernstein Polynomials

指導教授 : 吳裕振

摘要


本論文中我們對於現狀數據利用比例勝算比模型(詳見Bennett(1983a,b)) 來進行研究, 在柯興杰碩士論文(2011) 中, 他利用最大概似估計的方法而且使用比例勝算比模型, 研究現狀數據概似比檢定統計量。在倪裕程(2012) 的論文則是用貝氏方法來去進行估計, 但只對比例勝算比模型做估計, 且提出一個伯氏多項式應用於貝氏存活分析中, 而本篇論文其模型的假設和倪裕程(2012) 的論文是一樣的, 但估計參數方面, 們是用最大概似估計的方法。其模型無母數參數部分,一樣用伯氏多項式, 而在共變量Z 我們選擇離散僅0 或1的值,用馬可夫鏈蒙地卡羅法來尋找參數最大概似估計(Maximum Likelihood Estimator)。而 且在模擬計算中, 我們有不錯的表現。

並列摘要


In this paper, we use the status quo data proportional odds model (see Bennett (1983a, b)) to carry out research in the Ke Xingjie’s Thesis (2011), he used the maximum likelihood estimation method and the use of proportional odds ratio model, research data likelihood ratio test statistic. In Ni Yu Cheng’s (2012) paper is to come and go with the Bayesian approach to estimate, but only estimated the proportional odds model, and propose a Bernstein polynomial used in Bayesian survival analysis, this thesis the model assumptions and Ni Yu Cheng (2012) paper is the same, but the estimated parameters, we are using the maximum likelihood estimation method. The nonparametric parameters of the model part, as with Bernstein polynomials in the covariate Z discrete value of 0 or 1, with the Markov chain Monte Carlo method to find the parameters of maximum likelihood estimation (MLE). And simulation, we have a good performance.

參考文獻


[11] 倪裕程, 貝氏對現狀數據勝算比模型之研究, 中原大學碩士論文, 2012.
[12] 賴京鈺, 存活現狀數據下風險函數之最大概似估計, 中原大學碩士論文, 2011
[9] W. Q. Fang “ Bayesian Survival Analysis for Current Status Data ”. Department of Applied
the analysis of current status data” , Journal of th American Statistical Association 1996.
[6] I.S. Chang , C.A. Hsiung , Y.J. Wu , C.C. Yang “ Bayesian Survival Analysis Using

延伸閱讀