...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Quality Control and Applied Statistics >On the bias of the multiple-imputation variance estimator in survey sampling
【24h】

On the bias of the multiple-imputation variance estimator in survey sampling

机译：调查抽样中多输入方差估计量的偏差

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

To consider the applicability of the multiple imputation (MI) variance estimator for general complex survey sampling schemes under a superpopulation model in which the finite survey population is assumed to be a random sample from an infinite population. Imputation is widely used in sample surveys to assign values for item nonresponses. If the imputed values are ti'eated as if they were observed, then estimates of the variances of the estimates will generally be underestimates. Multiple imputation (MI) has been proposed as a method for estimating the precision of sample estimates in the presence of imputed values shown in Rubin (Refs. 1,2). MI is applied to a data set with missing items by repeating the process of assigning values for each of the missing values M times, creating M completed data sets. Each of the M completed data sets can be used to compute an estimate θ_I(k) of a popu,1ation parameter θ(k = 1,2,... M), with the subscript I indicating that some values have been imputed. Rubin (Ref. 1) proposed that θ be estimated by the average of the M estimates
机译：考虑多重插补（MI）方差估计量在超级人口模型下的一般复杂调查抽样方案的适用性，在该模型中，有限调查人口被视为来自无限人口的随机样本。插补被广泛用于样本调查中，以为项目不答复分配值。如果推定值像观察到的那样被处理，则估计值的方差估计值通常会被低估。提出了多重插补（MI）作为在鲁宾中显示的估算值存在的情况下估算样本估算精度的方法（参考文献1,2）。通过重复M次为每个缺失值分配值的过程，将MI应用于具有缺失项的数据集，从而创建M个完整的数据集。 M个完成的数据集中的每个数据集均可用于计算人口参数θ（k = 1,2，... M）的估计值θ_I（k），下标I表示已推算了一些值。鲁宾（参考文献1）提出用M个估计的平均值来估计θ

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号