本文已被：浏览 1061次 下载 542次
投稿时间：2014-02-27
投稿时间：2014-02-27
中文摘要: 概率密度图形非对称时,取对称分位点所确定的置信区间长度一般不是最短的,对于相同的置信水平,利用解方程组的方法可以求出单个正态总体中方差的最短置信区间及两个正态总体方差之比的最短置信区间,置信区间长度越短则估计的精度越高.
Abstract:When the graph of probability density is not symmetrical,the length of confidence interval given by symmetrical quantile is not always the shortest.For the same confidence level,the shortest of the confidence interval of the variance of a single normal distribution and the ratio of the variance of two normal distribution can be given by solving linear equations,and the shorter length of confidence interval means a higher precision of estimation.
文章编号：20140220 中图分类号： 文献标志码：
基金项目:
| 作者 | 单位 | |
| 蒋书法 | 上海电力学院数理学院 | DYJSF@163.com. |
| Author Name | Affiliation | |
| JIANG Shufa | School of Mathematics and Physics, Shanghai University of Electric Power, Shanghai 200090, China | DYJSF@163.com. |
引用文本：
蒋书法.最短置信区间的求解[J].上海电力大学学报,2014,30(2):188-192.
JIANG Shufa.Solution to the Shortest Confidence Interval[J].Journal of Shanghai University of Electric Power,2014,30(2):188-192.
蒋书法.最短置信区间的求解[J].上海电力大学学报,2014,30(2):188-192.
JIANG Shufa.Solution to the Shortest Confidence Interval[J].Journal of Shanghai University of Electric Power,2014,30(2):188-192.