首页|不同损失函数下Lindley分布参数的Bayes估计

不同损失函数下Lindley分布参数的Bayes估计

扫码查看
在熵损失函数和Q对称熵损失函数下,对参数的先验分布选取无信息先验分布和伽玛分布,研究了 Lindley分布参数的Bayes估计问题,且通过随机模拟比较不同条件下参数的Bayes估计效果.结果表明:同一种损失函数下,参数的先验分布为伽玛分布时估计效果更佳;样本容量较少时,在熵损失函数下,且先验分布为伽玛分布时,Bayes估计的均方误差较小;样本容量较多时,在Q对称熵损失函数及先验分布取伽玛分布的条件下,估计效果更理想.最后,由实例表明估计效果与数值模拟相符.
Bayes Estimation of Lindley Distribution Parameters under Different Loss Functions
Under the entropy loss function and Q-symmetric entropy loss function,the information-less prior distribution and gamma distribution are selected for the prior distribution of the parameters,eand the Bayes estimation problem of Lindley distribution parameters is studied,and the Bayes estima-tion effect of the parameters under different conditions is compared by random simulation.The results show that we conclude that under the same loss function,the estimation effect is better when the prior distribution of the parameters is the gamma distribution.When the sample size is small,under the en-tropy loss function,and the prior distribution is gamma distribution,Bayes'estimated mean squared er-ror is small;When the sample size is large,the estimation effect is more ideal under the condition of Q-symmetric entropy loss function and gamma distribution of prior distribution.Finally,examples show that the estimated effect matches the numerical simulation.

Lindley distributionentropy loss functionQ-symmetric entropy loss functionBayes estimates

赵孟茹、周菊玲

展开 >

新疆师范大学数学科学学院,新疆乌鲁木齐 830017

Lindley分布 熵损失函数 Q对称熵损失函数 Bayes估计

国家自然科学基金项目新疆师范大学教学研究与改革项目新疆师范大学科研发展专项项目

11801488SDJG2020-30XJNUZX202001

2024

淮阴师范学院学报(自然科学版)
淮阴师范学院

淮阴师范学院学报(自然科学版)

影响因子:0.259
ISSN:1671-6876
年,卷(期):2024.23(3)
  • 7