...
  • 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 >Reducing sample size needed for accelerated failure time model using more efficient sampling methods
【24h】

Reducing sample size needed for accelerated failure time model using more efficient sampling methods

机译：Reducing sample size needed for accelerated failure time model using more efficient sampling methods

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

摘要

It is important to use a cost effective sampling method that can yield more efficient estimators with less cost. Extreme ranked set sampling (ERSS) is a sampling techniques that has potential for more efficient estimators with less cost compared to simple random sampling (SRS). A more efficient survival regression analysis method is proposed for accelerated failure time (AFT) models based on an auxiliary variable known to be associated with the response variable. Parameter estimation is studied based on the maximum likelihood approach to provide an expression for the estimated variance based on an inverse information matrix. The asymptotic behavior of the maximum likelihood estimator is also studied. Performance of the proposed method is evaluated using simulation studies and a real data set. Use of ERSS increases the power significantly when implemented in AFT model settings. Also, ERSS is able to provide more efficient estimates of parameters associated with hazards ratios in terms of smaller mean square errors and narrower confidence intervals when compared to those under SRS.

著录项

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