摘要
主要讨论了当寿命分布是威布尔分布时删失数据的贝叶斯统计分析方法.在考虑尺度参数先验取为逆伽玛分布而形状参数先验分别取为离散分布和均匀分布条件下给出了多种删失数据场合参数的贝叶斯估计;同时为使得计算更为简便,给出了计算贝叶斯估计的Gibbs抽样方法.模拟结果表明给出的方法是有效可行的.
This paper discusses the Bayesian statistical analysis for censored life data from Weibull distribution. Several Bayesian Estimation methods are proposed when the prior of the scale parameter follows the inverted Gamma distribution and the shape pa- rameter follows a discrete distribution or a uniform distribution. The Gibbs sampling iteration algorithm is applied to simplify the computation of the Bayesian analysis. A numerical simulation is also given to show that the proposed method is efficient.
出处
《上海师范大学学报（自然科学版）》
2008年第1期28-34,共7页
Journal of Shanghai Normal University(Natural Sciences)
基金
国家自然科学基金(10571057)
上海高校选拔培养优秀青年教师科研专项基金(0123-01)
关键词
威布尔分布
删失数据
贝叶斯估计
Gibbs抽样方法
Weibull distribution
censored data
bayesian estimation
gibbs sampling iteration algorithm
