右設限和現狀數據混合資料下勝算比之研究

Odds rate estimation for mixed right-censored and current status data

徐誌延(Chih-Yen Hsu)

指導教授 ： 吳裕振

中原大學/理學院/應用數學研究所/碩士(2015年)

https://doi.org/10.6840/cycu201500150

摘要

本篇論文對右設限資料和現狀數據混合之存活分析，在王天佑(2015)提出貝氏的方法針對分佈函數作估計，而我們也對同樣資料型態，對致死率與存活率的比值(勝算比)作研究估計，用最大概似方法來估計分析，其中勝算比用伯氏多項式來描述，而參數較為複雜，使用馬可夫鏈蒙地卡羅法(M.C.M.C)之Simulated Annealing來計算。而且在模擬計算中, 我們有不錯的表現。

關鍵字

右設限 ； 最大概似估計 ； 勝算比 ； 現狀數據

並列摘要

The study aims to the survival analysis for mixed right-censored and current status data. Tian-yo Wang (2015) proposed Bayesian approach to estimate distribution function. We estimated the ratio of fatality rate and survival rate based on Wang’s proposition due to the similar data. The maximum likelihood estimation method was used to estimate and analyze the data. Odds rate was described by Bernstein polynomials. The parameters was calculated by using Markov chain Monte Carlo method (M.C.M.C)-Simulated Annealing because it was more complicated. And simulation, we have a good performance.

並列關鍵字

right-censored ； maximum likelihood estimate ； odds ratio ； current status data

參考文獻

Department of Applied Mathematic, Chung Yuan Christian University,

[7 ] 許時淮 (2015) 現狀數據資料下勝算比之研究，初稿

[1 ]W. Q. Fang ” Bayesian Survival Analysis for Current Status Data ”.

master thesis 2007.

[3 ]G. Casella , R.L. Berger ” Statistical inference ” , Duxbury Press 1990.