首页 | 本学科首页   官方微博 | 高级检索  
     检索      

Bootstrap在零膨胀泊松回归模型检验中的探索
引用本文:陈珊萍.Bootstrap在零膨胀泊松回归模型检验中的探索[J].科学技术与工程,2007,7(19):4821-4824.
作者姓名:陈珊萍
作者单位:河海大学常州校区数理教学部,常州,213022
摘    要:根据零膨胀计数数据的特点,将Bootstrap统计技术运用到零膨胀泊松回归模型的置信区间计算及参数假设检验。通过计算机模拟,得到:第一,当样本量较小时,Bootstrap对参数估计进行假设检验的结果更为保守;第二,Bootstrap的学生氏置信区间优于零膨胀泊松回归模型的区间估计。

关 键 词:Bootstrap  零膨胀泊松回归  泊松分布
文章编号:1671-1819(2007)19-4821-04
修稿时间:2007-06-13

Exploration of Bootstrap Inflated Poisson in the Testing of the Zero Regression Model
CHEN Shan-ping.Exploration of Bootstrap Inflated Poisson in the Testing of the Zero Regression Model[J].Science Technology and Engineering,2007,7(19):4821-4824.
Authors:CHEN Shan-ping
Institution:Department of Mathematics and Physics, Hohai University, Changzhou 213022, P. R. China
Abstract:Bootstrap is applied to the Zero Inflated Poisson regression model in confidence intervals and hypothesis testing with Zero Inflated data. The outcomes of the hypothesis testing from the Bootstrap method are more conservative than the ones from the ZIP regression model when sample sizes are small. Also the confidence intervals from the Bootstrap method are more accurate than the ones from the ZIP regression model.
Keywords:Bootstrap  ZIP regression  Poisson distribution
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《科学技术与工程》浏览原始摘要信息
点击此处可从《科学技术与工程》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号