
构造泊松均值固定宽度置信区间的序贯方法和两阶段方法
Fixed-Length Confidence Intervals for the Poisson Mean via Sequential Methods and Two-Stage Methods
对于泊松分布的未知参数, 为了确定在构造指定覆盖率的固定宽度置信区间时的停止规则, 文章首先给出了一般序贯方法和两阶段方法的具体过程. 更进一步地, 分别基于均值估计量的精确分布和等价分布提出了两种新的停止规则, 并给出了序贯和两阶段停止时间的渐近性质. 最后, 通过蒙特卡罗模拟对于所提方法进行了比较, 并进行了实证分析.
To determine stopping rules in the process of obtaining fixed-width confidence intervals of prescribed coverage probability for the unknown mean parameter of a Poisson distribution, this article firstly gives the specific procedure of a general sequential method, as well as a two-stage procedure. Furthermore, two new stopping rules are derived respectively based on the exact and equivalent distribution of the mean estimator, with the asymptotic properties of the sequential and two-stage stopping time demonstrated. Finally, we illustrate the simulations for all the proposed methods for comparisons and conduct a real data analysis.
固定宽度置信区间 / / 泊松分布 / / 序贯方法 / / 两阶段抽样. {{custom_keyword}} /
/
〈 |
|
〉 |