具有缺失数据的Eichhorn模型

Missing Data in Eichhorn Model

为了解决在Eichhorn乘法扰动模型中存在的项目无回答问题,对敏感变量总体均值在辅助变量总体均值已知与未知条件下提出了比率插补方法.理论比较和数值模拟得出的结果表明提出的插补方法比传统的方法效率更高.

To deal with the problem of Non-response in Eichhorn multiplicative scrambled model, the ratio imputation with complete or incomplete auxiliary variable is addressed to estimate the population mean of sensitive characteristic variable. Theoretically, the mean square error （MSE） of all proposed estimators are obtained. And it is shown that the proposed estimators are always more efficient than the traditional estimators. The merits of the suggested estimators have been examined through a numerical illustration.