摘要
对枢轴量G的分布具有单峰密度函数的情形,证明了G的最短置信区间是满足置信区间端点密度函数值"等高"条件的置信区间.还对枢轴量分布为正态分布、t分布、卡方分布、F分布、伽玛分布和对数正态分布情形下,在Excel中进行了搜索式数值计算,并列表比较了"等尾"和"等高"情形下置信区间的长度,验证了上述分析结论.另外,还讨论了枢轴量的最短置信区间与枢轴量中所含参数的最短置信区间的关系.
This paper proves that the shortest confidence interval of the pivot with unimodal density function is to keep the confidence interval endpoint density function values "equal".The paper also conducts a numerical calculations in Excel and compares the confidence interval lengths in two different situations for the pivot follows normal distribution,t distribution,chisquare distribution,F distribution,gamma distribution and logarithmic normal.The results confirm the above analysis conclusions.In addition,the paper also discusses the relationship between the shortest confidence interval of the pivot and the shortest confidence interval parameters contained in the pivot.
引文
[1]郭民之.概率统计实验[M].北京:北京大学出版社,2012.
[2]张庆平.非对称分布置信区间的分析[J].统计与决策,2007,5(2):137.
[3]何宾.区间估计的一种求解方法[J].统计与决策,2005,7(2):139.
[4]韦博成.参数统计教程[M].北京:高等教育出版社,2006.
[5]茆诗松,王静龙,襥晓龙.高等数理统计(第二版)[M].北京:高等教育出版社,2006
[6]Sheldon M.Ross.Simulation(4th)[M].Elsevier,2006.