...
  • 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 modelling and estimation for the relative risk and risk difference
【24h】

On modelling and estimation for the relative risk and risk difference

机译：关于相对风险与风险差异的建模与估算

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

Though odds ratio is the most common way to model the association between binary random variables, there is a problem in interpreting odds ratios mainly due to noncollapsibility. Relative risk (RR) and the risk difference (RD) are often treated as alternative measures that are collapsible and are simpler to interpret. There are problems in using the generalized linear model for modeling the dependence of RR or RD. In the article this problem is addressed by proposing the conditional log odds-product as a preferred nuisance model that represents the variation dependence between these risk parameters (RR and RD) and the baseline risk. The proposed novel nuisance model not only facilitates maximum- likelihood estimation but also permits doubly-robust estimation for the parameters of interest. In order to illustrate the proposed approach, both simulations and real time data analyses are used. The proposed general approach to modeling the RR and RD as functions of baseline covariates fills the methodological gap.
机译：虽然赔率比是模拟二进制随机变量之间的关联的最常见方法，但在解释差距之比上主要是由于不可渗透率的问题。相对风险（RR）和风险差异（RD）通常被视为可收回的替代措施，更简单地解释。使用广义线性模型有问题用于建模RR或RD的依赖性。在本文中，通过提出条件日志赔率 - 产品作为优选的滋扰模型来解决该问题，该模型代表这些风险参数（RR和RD）与基线风险之间的变化依赖性。所提出的新型滋扰模型不仅有助于最大似然估计，而且还允许对感兴趣的参数进行双重稳健的估计。为了说明所提出的方法，使用仿真和实时数据分析。提出的一般方法为基线协变量的职能建模RR和RD填补了方法论差距。

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