数学物理学报 ›› 2021, Vol. 41 ›› Issue (1): 194-216.

• 论文 • 上一篇    下一篇

多个偏正态总体共同位置参数的Bootstrap置信区间

叶仁道1,*(),王仲池1,罗堃2,林雅1   

  1. 1 杭州电子科技大学经济学院 杭州 310018
    2 杭州师范大学阿里巴巴商学院 杭州 310036
  • 收稿日期:2019-11-22 出版日期:2021-02-26 发布日期:2021-01-29
  • 通讯作者: 叶仁道 E-mail:yerendao2003@163.com
  • 基金资助:
    教育部人文社会科学研究项(19YJA910006);浙江省自然科学基金(LY20A010019);浙江省属高校基本科研业务费专项资金(GK199900299012-204);国家自然科学基金(11401148)

Bootstrap Confidence Intervals for the Common Location Parameter of Several Skew-Normal Populations

Rendao Ye1,*(),Zhongchi Wang1,Kun Luo2,Ya Lin1   

  1. 1 School of Economics, Hangzhou Dianzi University, Hangzhou 310018
    2 Alibaba Business College, Hangzhou Normal University, Hangzhou 310036
  • Received:2019-11-22 Online:2021-02-26 Published:2021-01-29
  • Contact: Rendao Ye E-mail:yerendao2003@163.com
  • Supported by:
    the Humanities and Social Science Projects of the Ministry of Education(19YJA910006);the NSF of Zhejiang Province(LY20A010019);the Fundamental Research Funds for the Provincial Universities of Zhejiang(GK199900299012-204);NSFC(11401148)

摘要:

该文基于Bootstrap方法研究多个偏正态总体共同位置参数的区间估计和假设检验问题.首先,分别给出未知参数的矩估计和极大似然估计.其次,将徐礼文[1]对多个正态总体共同均值的探讨推广到多个偏正态总体,进而构造共同位置参数的Bootstrap置信区间和Bootstrap检验统计量. Monte Carlo模拟结果表明,无论是两个总体、三个总体还是五个总体,基于矩估计和惩罚极大似然估计的Bootstrap置信区间在覆盖概率意义下优于其他四种Bootstrap置信区间.最后,将上述方法应用于地区生产总值和生物利用度数据的案例分析,以验证该文所给方法的合理性和有效性.

关键词: 偏正态总体, 共同位置参数, 矩估计, 极大似然估计, Bootstrap置信区间

Abstract:

In this paper, we consider the interval estimation and hypothesis testing problems for the common location parameter of several skew-normal populations when the scale parameters and skewness parameters are unknown. Firstly, we estimate the unknown parameters using the methods of moment estimation and maximum likelihood estimation. Secondly, the Bootstrap confidence intervals and Bootstrap test statistics are constructed, which generalize the results given by Xu [1] under several normal populations. Thirdly, the Monte Carlo simulation results indicate that the Bootstrap confidence intervals based on the methods of moment estimation and maximum penalized likelihood estimation perform better than other four confidence intervals. Finally, the above approaches are applied to the real data examples of regional gross domestic product of China and bioavailability in order to verify the reasonableness and effectiveness of the proposed approaches.

Key words: Skew-normal population, Common location parameter, Moment estimation, Maximum likelihood estimation, Bootstrap confidence interval

中图分类号: 

  • O212