Search

We propose two novel approaches for estimating time-varying effects of functional predictors within a linear functional Cox model framework. This model allows for time-varying associations of a functional predictor observed at baseline, estimated using penalized regression splines for smoothness across the functional domain and event time. The first approach, suitable for small-to-medium datasets, uses the Cox-Poisson likelihood connection for valid estimation and inference. The second, a landmark approach, significantly reduces computational burden for large datasets and high-dimensional functional predictors. Both methods address proportional hazards violations for functional predictors and model associations as a bivariate smooth coefficient. Motivated by analyzing diurnal motor activity patterns and all-cause mortality in NHANES (N=4445, functional predictor dimension=1440), we demonstrate the first method's computational limitations and the landmark approach's efficiency. These methods are implemented in stable, high-quality software using the mgcv package for penalized spline regression with automated smoothing parameter selection. Simulations show both methods achieve high accuracy in estimating functional coefficients, with the landmark approach being computationally faster but slightly less accurate. The Cox-Poisson method provides nominal coverage probabilities, while landmark inference was not assessed due to inherent bias. Sensitivity to landmark modeling choices was evaluated. Application to NHANES reveals an attenuation of diurnal effects on mortality over an 8-year follow-up.

Dynamic prediction, which typically refers to the prediction of future outcomes using historical records, is often of interest in biomedical research. For datasets with large sample sizes, high measurement density, and complex correlation structures, traditional methods are often infeasible because of the computational burden associated with both data scale and model complexity. Moreover, many models do not directly facilitate out-of-sample predictions for generalized outcomes. To address these issues, we develop a novel approach for dynamic predictions based on a recently developed method estimating complex patterns of variation for exponential family data: fast Generalized Functional Principal Components Analysis (fGFPCA). Our method is able to handle large-scale, high-density repeated measures much more efficiently with its implementation feasible even on personal computational resources (e.g., a standard desktop or laptop computer). The proposed method makes highly flexible and accurate predictions of future trajectories for data that exhibit high degrees of nonlinearity, and allows for out-of-sample predictions to be obtained without reestimating any parameters. A simulation study is designed and implemented to illustrate the advantages of this method. To demonstrate its practical utility, we also conducted a case study to predict diurnal active/inactive patterns using accelerometry data from the National Health and Nutrition Examination Survey (NHANES) 2011-2014. Both the simulation study and the data application demonstrate the better predictive performance and high computational efficiency of the proposed method compared to existing methods. The proposed method also obtains more personalized prediction that improves as more information becomes available, which is an essential goal of dynamic prediction that other methods fail to achieve.

We propose generalized conditional functional principal components analysis (GC-FPCA) for the joint modeling of the fixed and random effects of non-Gaussian functional outcomes. The method scales up to very large functional data sets by estimating the principal components of the covariance matrix on the linear predictor scale conditional on the fixed effects. This is achieved by combining three modeling innovations: (1) fit local generalized linear mixed models (GLMMs) conditional on covariates in windows along the functional domain; (2) conduct a functional principal component analysis (FPCA) on the person-specific functional effects obtained by assembling the estimated random effects from the local GLMMs; and (3) fit a joint functional mixed effects model conditional on covariates and the estimated principal components from the previous step. GC-FPCA was motivated by modeling the minute-level active/inactive profiles over the day ($1{,}440$ 0/1 measurements per person) for $8{,}700$ study participants in the National Health and Nutrition Examination Survey (NHANES) 2011-2014. We show that state-of-the-art approaches cannot handle data of this size and complexity, while GC-FPCA can., Comment: 38 pages with supplementary material, 5 figures for the main article and 4 supplementary figures, 3 tables for the main article and 10 supplementary tables, submitted to Journal of Computational and Graphical Statistics (JCGS)

Predicting the timing and occurrence of events is a major focus of data science applications, especially in the context of biomedical research. Performance for models estimating these outcomes, often referred to as time-to-event or survival outcomes, is frequently summarized using measures of discrimination, in particular time-dependent AUC and concordance. Many estimators for these quantities have been proposed which can be broadly categorized as either semi-parametric estimators or non-parametric estimators. In this paper, we review various estimators' mathematical construction and compare the behavior of the two classes of estimators. Importantly, we identify a previously unknown feature of the class of semi-parametric estimators that can result in vastly over-optimistic out-of-sample estimation of discriminative performance in common applied tasks. Although these semi-parametric estimators are popular in practice, the phenomenon we identify here suggests this class of estimators may be inappropriate for use in model assessment and selection based on out-of-sample evaluation criteria. This is due to the semi-parametric estimators' bias in favor of models that are overfit when using out-of-sample prediction criteria (e.g., cross validation). Non-parametric estimators, which do not exhibit this behavior, are highly variable for local discrimination. We propose to address the high variability problem through penalized regression splines smoothing. The behavior of various estimators of time-dependent AUC and concordance are illustrated via a simulation study using two different mechanisms that produce over-optimistic out-of-sample estimates using semi-parametric estimators. Estimators are further compared using a case study using data from the National Health and Nutrition Examination Survey (NHANES) 2011-2014.

Between 2011 and 2014 NHANES collected objectively measured physical activity data using wrist-worn accelerometers for tens of thousands of individuals for up to seven days. Here we analyze the minute-level indicators of being active, which can be viewed as binary (because there is an active indicator at every minute), multilevel (because there are multiple days of data for each study participant), functional (because within-day data can be viewed as a function of time) data. To extract within- and between-participant directions of variation in the data, we introduce Generalized Multilevel Functional Principal Component Analysis (GM-FPCA), an approach based on the dimension reduction of the linear predictor. Scores associated with specific patterns of activity are shown to be strongly associated with time to death. Extensive simulation studies indicate that GM-FPCA provides accurate estimation of model parameters, is computationally stable, and is scalable in the number of study participants, visits, and observations within visits. R code for implementing the method is provided.

We consider the problem of predicting an individual's identity from accelerometry data collected during walking. In a previous paper we introduced an approach that transforms the accelerometry time series into an image by constructing its complete empirical autocorrelation distribution. Predictors derived by partitioning this image into grid cells were used in logistic regression to predict individuals. Here we: (1) implement machine learning methods for prediction using the grid cell-derived predictors; (2) derive inferential methods to screen for the most predictive grid cells; and (3) develop a novel multivariate functional regression model that avoids partitioning of the predictor space into cells. Prediction methods are compared on two open source data sets: (1) accelerometry data collected from $32$ individuals walking on a $1.06$ kilometer path; and (2) accelerometry data collected from six repetitions of walking on a $20$ meter path on two separate occasions at least one week apart for $153$ study participants. In the $32$-individual study, all methods achieve at least $95$% rank-1 accuracy, while in the $153$-individual study, accuracy varies from $41$% to $98$%, depending on the method and prediction task. Methods provide insights into why some individuals are easier to predict than others., Comment: 37 pages, 6 figures, 2 tables. Submitted to Journal of the American Statistical Association

We propose a new fast generalized functional principal components analysis (fast-GFPCA) algorithm for dimension reduction of non-Gaussian functional data. The method consists of: (1) binning the data within the functional domain; (2) fitting local random intercept generalized linear mixed models in every bin to obtain the initial estimates of the person-specific functional linear predictors; (3) using fast functional principal component analysis to smooth the linear predictors and obtain their eigenfunctions; and (4) estimating the global model conditional on the eigenfunctions of the linear predictors. An extensive simulation study shows that fast-GFPCA performs as well or better than existing state-of-the-art approaches, it is orders of magnitude faster than existing general purpose GFPCA methods, and scales up well with both the number of observed curves and observations per curve. Methods were motivated by and applied to a study of active/inactive physical activity profiles obtained from wearable accelerometers in the NHANES 2011-2014 study. The method can be implemented by any user familiar with mixed model software, though the R package fastGFPCA is provided for convenience.

Linear mixed-effects models are widely used in analyzing repeated measures data, including clustered and longitudinal data, where inferences of both fixed effects and variance components are of importance. Unlike the fixed effect inference that has been well studied, inference on the variance components is more challenging due to null value being on the boundary and the nuisance parameters of the fixed effects. Existing methods often require strong distributional assumptions on the random effects and random errors. In this paper, we develop empirical likelihood-based methods for the inference of the variance components in the presence of fixed effects. A nonparametric version of the Wilks' theorem for the proposed empirical likelihood ratio statistics for variance components is derived. We also develop an empirical likelihood test for multiple variance components related to a sequence of correlated outcomes. Simulation studies demonstrate that the proposed methods exhibit better type 1 error control than the commonly used likelihood ratio tests when the Gaussian distributional assumptions of the random effects are violated. We apply the methods to investigate the heritability of physical activity as measured by wearable device in the Australian Twin study and observe that such activity is heritable only in the quantile range from 0.375 to 0.514.

Continuous glucose monitors (CGMs) are increasingly used to measure blood glucose levels and provide information about the treatment and management of diabetes. Our motivating study contains CGM data during sleep for 174 study participants with type II diabetes mellitus measured at a 5-minute frequency for an average of 10 nights. We aim to quantify the effects of diabetes medications and sleep apnea severity on glucose levels. Statistically, this is an inference question about the association between scalar covariates and functional responses. However, many characteristics of the data make analyses difficult, including (1) non-stationary within-day patterns; (2) substantial between-day heterogeneity, non-Gaussianity, and outliers; 3) large dimensionality due to the number of study participants, sleep periods, and time points. We evaluate and compare two methods: fast univariate inference (FUI) and functional additive mixed models (FAMM). We introduce a new approach for calculating p-values for testing a global null effect of covariates using FUI, and provide practical guidelines for speeding up FAMM computations, making it feasible for our data. While FUI and FAMM are philosophically different, they lead to similar point estimators in our study. In contrast to FAMM, FUI is fast, accounts for within-day correlations, and enables the construction of joint confidence intervals. Our analyses reveal that: (1) biguanide medication and sleep apnea severity significantly affect glucose trajectories during sleep, and (2) the estimated effects are time-invariant.

Background: We propose a method for estimating the timing of in-bed intervals using objective data in a large representative U.S. sample, and quantify the association between these intervals and age, sex, and day of the week. Methods: The study included 11,951 participants six years and older from the National Health and Nutrition Examination Survey (NHANES) 2003-2006, who wore accelerometers to measure physical activity for seven consecutive days. Participants were instructed to remove the device just before the nighttime sleep period and put it back on immediately after. This nighttime period of non-wear was defined in this paper as the objective bedtime (OBT), an objectively estimated record of the in-bed-interval. For each night of the week, we estimated two measures: the duration of the OBT (OBT-D) and, as a measure of the chronotype, the midpoint of the OBT (OBT-M). We estimated day-of-the-week-specific OBT-D and OBT-M using gender-specific population percentile curves. Differences in OBT-M (chronotype) and OBT-D (the amount of time spent in bed) by age and sex were estimated using regression models. Results: The estimates of OBT-M and their differences among age groups were consistent with the estimates of chronotype obtained via self-report in European populations. The average OBT-M varied significantly by age, while OBT-D was less variable with age. The most pronounced differences were observed between OBT-M of weekday and weekend nights. Conclusions: The proposed measures, OBT-D and OBT-M, provide useful information of time in bed and chronotype in NHANES 2003-2006. They identify within-week patterns of bedtime and can be used to study associations between the bedtime and the large number of health outcomes collected in NHANES 2003-2006.