pammtools: Piece-wise exponential Additive Mixed Modeling tools
pammtools: Piece-wise exponential Additive Mixed Modeling tools
复制标题
pammtools：分段指数加法混合建模工具
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
F. Scheipl
中科院分区：
文献类型：
--
作者：
Andreas Bender;F. Scheipl
本文介绍了 pammtools 包，它有助于分段指数加法混合模型的数据转换、估计和解释。特别关注的是随时间变化的效应和时间相关协变量的累积效应，其中过去对协变量的多次观察可以累积影响风险，可能由非线性函数加权。该软件包提供了方便模拟和可视化此类效应的功能，以及将事件时间数据从标准格式转换为适合其估计的格式的强大且多功能的功能。这些模型可以表示为广义加性混合模型，并使用 R 包 mgcv 进行估计。整篇文章提供了许多有关真实数据和模拟数据的示例以及相应的 R 代码。
This article introduces the pammtools package, which facilitates data transformation, estimation and interpretation of Piece-wise exponential Additive Mixed Models. A special focus is on time-varying effects and cumulative effects of time-dependent covariates, where multiple past observations of a covariate can cumulatively affect the hazard, possibly weighted by a non-linear function. The package provides functions for convenient simulation and visualization of such effects as well as a robust and versatile function to transform time-to-event data from standard formats to a format suitable for their estimation. The models can be represented as Generalized Additive Mixed Models and estimated using the R package mgcv. Many examples on real and simulated data as well as the respective R code are provided throughout the article.