逆抽样条件下相对差的估计  

Estimation of Relative Difference under Inverse Sampling

作　　者：吴武琴[1] 左艳芳[1] 李会琼[2] 江绍萍[3] 

机构地区：[1]昆明冶金高等专科学校社会科学与公共学院,云南昆明650033 [2]云南大学数学与统计学院,云南昆明650091 [3]云南民族大学数学与计算机科学学院,云南昆明650050

出　　处：《昆明理工大学学报（理工版）》2009年第5期108-111,124,共5页Journal of Kunming University of Science and Technology(Natural Science Edition)

基　　金：云南大学理(工)科校级科研项目资助(项目编号:2008YB025);云南省教育厅科学研究基金项目(项目编号:07L00001)

摘　　要：在连续抽样过程中,当我们感兴趣的样本出现的概率很小时,逆抽样比一般的二项分布更具有实用性.论文中建立了Wald统计量和对数Wald统计量,并利用Filler定理求得两个统计量的感兴趣参数的置信区间,且基于统计量的条件下计算了覆盖概率和平均长度.通过模拟研究,发现对数Wald统计量的参数的置信区间能够保证在覆盖概率大的条件下的平均长度也较小.实例分析表明,该方法是有效的.Inverse sampling is considered to be a more appropriate sampling scheme than the usual binomial sampling scheme when the subjects arrive sequentially and when the underlying response of interest is acute. In this paper, we study various test statistics for testing relative difference in case -control studies under inverse sam- pling, which include the Wald and logarithm Wald test statistics. Filler theorem is then employed to get the confidence interval of interest parameter. The estimated coverage probability and the average length are therefore obtained. Through simulation, it is found that logarithm Wald can get large coverage probability when there is a small average length. A real example is finally used to illustrate the proposed method.

关 键 词：逆抽样 相对差 Wald统计量 对数Wald统计量 

分 类 号：O213[理学—概率论与数理统计]