Modeling time-varying effects with generalized and unsynchronized longitudinal data.
Modeling time-varying effects with generalized and unsynchronized longitudinal data.
复制标题
DOI：
10.1002/sim.5740
复制
发表时间：
2013-07-30
影响因子：
2
通讯作者：
Nguyen, Danh V.
中科院分区：
文献类型：
--
作者：
Sentuerk, Damla;Dalrymple, Lorien S.;Mohammed, Sandra M.;Kaysen, George A.;Nguyen, Danh V.
关键词：
我们提出了新的广义变系数模型，为不同步，不规则和罕见的纵向设计/数据量身定制的估计方法。非同步纵向数据是指在不同时间点测量的每个个体的时间依赖性响应和协变量测量。所提出的方法的动机是从综合透析研究（CDS）的数据。我们在透析开始后的前两年内，建立了感染相关住院状态与炎症标志物C反应蛋白（CRP）之间潜在的年龄变化相关性模型。传统的纵向建模不能直接应用于不同步的数据，并且不存在估计广义结果的时间或年龄变化效应的方法（例如，二进制或计数数据）。此外，通过对CDS数据的分析和模拟研究，我们发现，在这种情况下，同步数据以应用传统建模所需的预处理步骤（如分箱）可能会导致信息的重大损失。相比之下，所提出的方法不丢弃任何观察;它们利用这样一个事实，即虽然由于不规则性和不频繁性，单个受试者轨迹中的信息很少，但通过使用功能数据分析汇集来自所有受试者的信息，可以准确有效地恢复底层过程的时刻。受试者特定的平均响应轨迹的预测推导和有限样本性质的估计进行了研究。
We propose novel estimation approaches for generalized varying coefficient models that are tailored for unsynchronized, irregular and infrequent longitudinal designs/data. Unsynchronized longitudinal data refers to the time-dependent response and covariate measurements for each individual measured at distinct time points. The proposed methods are motivated by data from the Comprehensive Dialysis Study (CDS). We model the potential age-varying association between infection-related hospitalization status and the inflammatory marker, C-reactive protein (CRP), within the first two years from initiation of dialysis. Traditional longitudinal modeling cannot directly be applied to unsynchronized data and no method exists to estimate time- or age-varying effects for generalized outcomes (e.g., binary or count data) to date. In addition, through the analysis of the CDS data and simulation studies, we show that preprocessing steps, such as binning, needed to synchronize data to apply traditional modeling can lead to significant loss of information in this context. In contrast, the proposed approaches discard no observation; they exploit the fact that although there is little information in a single subject trajectory due to irregularity and infrequency, the moments of the underlying processes can be accurately and efficiently recovered by pooling information from all subjects using functional data analysis. Subject-specific mean response trajectory predictions are derived and finite sample properties of the estimators are studied.
登录
查看更多内容
影响因子：
2.7
作者：
Hoover, DR;Rice, JA;Yang, LP
通讯作者：
Yang, LP
影响因子：
3.7
作者：
Cai, ZW;Fan, JQ;Li, RZ
通讯作者：
Li, RZ
影响因子：
4.5
作者：
Yao, F;Müller, HG;Wang, JL
通讯作者：
Wang, JL
DOI：
10.2215/cjn.03090509
发表时间：
2009-12-01
影响因子：
9.8
作者：
Kaysen, George A.
通讯作者：
Kaysen, George A.
影响因子：
2.7
作者：
Huang, JHZ;Wu, CO;Zhou, L
通讯作者：
Zhou, L