Thumbnail Image
Item

Generalized functional linear model with a point process predictor

Sun, Jiehuan
; 
Lee, Kuang-Yao
Citations
Altmetric:
Genre
Journal article
Date
2024-02-08
Advisor
Committee member
Group
Department
Statistics, Operations, and Data Science
Subject
Generalized functional linear model
 Joint modeling
 Point process data
 Variational approximation
Permanent link to this record
https://scholarshare.temple.edu/handle/20.500.12613/11800
Collections
Faculty/Researcher Works
Show all metadata
Files
Thumbnail Image
Lee-JournalArticle-2024-02.pdf
Adobe PDF, 1.57 MB
Research Projects
Organizational Units
Journal Issue
DOI
https://doi.org/10.1002/sim.10023
Abstract
Point process data have become increasingly popular these days. For example,many of the data captured in electronic health records (EHR) are in the formatof point process data. It is of great interest to study the association between apoint process predictor and a scalar response using generalized functional lin-ear regression models. Various generalized functional linear regression modelshave been developed under different settings in the past decades. However, exist-ing methods can only deal with functional or longitudinal predictors, not pointprocess predictors. In this article, we propose a novel generalized functional lin-ear regression model for a point process predictor. Our proposed model is basedon the joint modeling framework, where we adopt a log-Gaussian Cox processmodel for the point process predictor and a generalized linear regression modelfor the outcome. We also develop a new algorithm for fast model estimationbased on the Gaussian variational approximation method. We conduct exten-sive simulation studies to evaluate the performance of our proposed method andcompare it to competing methods. The performance of our proposed method isfurther demonstrated on an EHR dataset of patients admitted into the intensivecare units of the Beth Israel Deaconess Medical Center between 2001 and 2008.
Description
Citation
Sun J, Lee K-Y. Generalized functional linear model with a point process predictor. Statistics in Medicine. 2024; 43(8): 1564-1576. doi: 10.1002/sim.10023
Citation to related work
Wiley
Has part
Statistics in Medicine, Vol. 43, Iss. 8
ADA compliance
Embedded videos
License
Attribution-NonCommercial-NoDerivs CC BY-NC-ND
cbn