Parametric likelihood inference with censored survival data under the COM-Poisson cure models
2017
對設限存活資料(censored survival data)分析,Rodrigues等(2009)提出用Conway-Maxwell-Poisson (COM-Poisson)分佈為治愈模型(cure rate model)。對COM-Poisson治愈模型之特例——伯努利治愈模型(Bernoulli cure rate model),考慮使用不同之運算演算法,以最大概似估計法(maximum likelihood estimation)得參數之估計值。據Balakrishnan與Pal於2016以韋伯分佈(Weibull distribution)及於2015以廣義伽瑪分佈(generalized gamma distribution),假設為其壽命分佈(lifetime distribution)。進而導出之評分函數(score function)與黑塞矩陣(Hessian matrix),用以牛頓-拉弗森演算法(Newton-Raphson algorithm)及最大期望演算法(EM algorithm)。模擬為分析比較此二種演算法之表現。末了,實際資料分析作詳加闡明此方法模型。
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI