分组截尾数据下离散型寿命概率分布的估计方法

摘要

At present, the grouped data to unknown survival discrete probability distributions are less studied. To a-void solving complex nonlinear maximum likelihood equations,the probability distribution formula with recursion relations is derived from the likelihood equations by joining the Lagrange multiplier. Then the maximum likelihood estimation of p1 is obtained through the degradation of single interval model. Thus , the probability distribution of pi is calculated successively. The experiment result shows that this method is effective.%目前针对离散型未知寿命分布的分组数据研究较少.为避免求解复杂的非线性极大似然方程组,由加入拉格朗日乘子的似然方程组中推导具有递推关系的概率分布计算公式,并通过退化的单区间模型得到p1的极大似然估计,从而完成概率分布律p1的递推计算.实验说明该方法有效.