摘要
研究了Ⅰ型逐阶删失数据下基于EM算法的Weibull参数估计,模拟产生不同Weibull参数组合和删失计划下的Ⅰ型逐阶删失数据,应用基于,EM算法的极大似然估计方法得到参数的估计值,并与数值方法得到的极大似然估计值进行对比,说明EM算法的估计效果.对73名肾脏移植患者生存数据进行实例分析,验证了基于EM算法的参数估计方法的可行性.
This paper studies the parameters estimation of Weibull based on em algorithm with progressive type-I censored data. We produce progressive type-I censored data by different parameters of Weibull and progressive schemes. Estimate the parameters using an EM algorithm based on maximum likelihood estimation method. Compared with the numerical method, EM algorithm has better estimation results. EM algorithm is used to simulate the real survival data of 73 patients with renal transplantation ,verify the validity of parameter estimation method based on EM algorithm.
出处
《数学的实践与认识》
北大核心
2015年第16期196-202,共7页
Mathematics in Practice and Theory
基金
北京市高校创新人才项目(201206206)
