Sentürk Damla, Ghosh Samiran, Nguyen Danh V
Department of Biostatistics, University of California, Los Angeles, CA, USA.
Department of Family Medicine and Public Health Sciences, Wayne State University School of Medicine, MI, USA ; Center for Molecular Medicine and Genetics, Wayne State University School of Medicine, MI, USA.
Comput Stat Data Anal. 2014 May 1;73:1-15. doi: 10.1016/j.csda.2013.11.001.
Motivated by a longitudinal study on factors affecting the frequency of clinic visits of older adults, an exploratory time varying lagged regression analysis is proposed to relate a longitudinal response to multiple cross-sectional and longitudinal predictors from time varying lags. Regression relations are allowed to vary with time through smooth varying coefficient functions. The main goal of the proposal is to detect deviations from a concurrent varying coefficient model potentially in a subset of the longitudinal predictors with nonzero estimated lags. The proposed methodology is geared towards irregular and infrequent data where different longitudinal variables may be observed at different frequencies, possibly at unsynchronized time points and contaminated with additive measurement error. Furthermore, to cope with the curse of dimensionality which limits related current modeling approaches, a sequential model building procedure is proposed to explore and select the time varying lags of the longitudinal predictors. The estimation procedure is based on estimation of the moments of the predictor and response trajectories by pooling information from all subjects. The finite sample properties of the proposed estimation algorithm are studied under various lag structures and correlation levels among the predictor processes in simulation studies. Application to the clinic visits data show the effect of cognitive and functional impairment scores from varying lags on the frequency of the clinic visits throughout the study.
受一项关于影响老年人门诊就诊频率因素的纵向研究的启发，本文提出了一种探索性的时变滞后回归分析方法，用于将纵向响应与来自时变滞后的多个横截面和纵向预测变量联系起来。回归关系可以通过平滑变化系数函数随时间变化。该方法的主要目标是检测与并发变化系数模型的偏差，这种偏差可能存在于具有非零估计滞后的纵向预测变量子集中。所提出的方法适用于不规则和不频繁的数据，其中不同的纵向变量可能以不同的频率被观测到，可能在不同步的时间点，并且受到加性测量误差的影响。此外，为了应对限制当前相关建模方法的维度诅咒问题，本文提出了一种顺序模型构建程序，以探索和选择纵向预测变量的时变滞后。估计过程基于通过汇总所有受试者的信息来估计预测变量和响应轨迹的矩。在模拟研究中，研究了所提出的估计算法在各种滞后结构和预测变量过程之间的相关水平下的有限样本性质。将该方法应用于门诊就诊数据，展示了不同滞后的认知和功能损害评分对整个研究期间门诊就诊频率的影响。
