A Bayesian semi‐parametric scalar‐on‐function regression with measurement error using instrumental variables
Open Access
- 8 July 2024
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 43 (21) , 4043-4054
Abstract
Wearable devices such as the ActiGraph are now commonly used in research to monitor or track physical activity. This trend corresponds with the growing need to assess the relationships between physical activity and health outcomes, such as obesity, accurately. Device‐based physical activity measures are best treated as functions when assessing their associations with scalar‐valued outcomes such as body mass index. Scalar‐on‐function regression (SoFR) is a suitable regression model in this setting. Most estimation approaches in SoFR assume that the measurement error in functional covariates is white noise. Violating this assumption can lead to underestimating model parameters. There are limited approaches to correcting measurement errors for frequentist methods and none for Bayesian methods in this area. We present a non‐parametric Bayesian measurement error‐corrected SoFR model that relaxes all the constraining assumptions often involved with these models. Our estimation relies on an instrumental variable allowing a time‐varying biasing factor, a significant departure from the current generalized method of moment (GMM) approach. Our proposed method also permits model‐based grouping of the functional covariate following measurement error correction. This grouping of the measurement error‐corrected functional covariate allows additional ease of interpretation of how the different groups differ. Our method is easy to implement, and we demonstrate its finite sample properties in extensive simulations. Finally, we applied our method to data from the National Health and Examination Survey to assess the relationship between wearable device‐based measures of physical activity and body mass index in adults in the United States.Keywords
Funding Information
- National Cancer Institute (U01‐CA057030‐29S1, U01‐CA057030‐29S2)
- National Institute of Diabetes and Digestive and Kidney Diseases (R01DK132385, 1R01DK136994‐01A1)
This publication has 22 references indexed in Scilit:
- Instrumental variable approach to estimating the scalar‐on‐function regression model with measurement error with application to energy expenditure assessment in childhood obesityStatistics in Medicine, 2019
- A tutorial on Dirichlet process mixture modelingJournal of Mathematical Psychology, 2019
- Shrinkage priors for Bayesian penalized regressionJournal of Mathematical Psychology, 2019
- Bayesian Semiparametric Multivariate Density DeconvolutionJournal of the American Statistical Association, 2017
- How valid are wearable physical activity trackers for measuring steps?European Journal of Sport Science, 2016
- Methods for Scalar‐on‐Function RegressionInternational Statistical Review, 2016
- Amount of Time Spent in Sedentary Behaviors in the United States, 2003-2004American Journal of Epidemiology, 2008
- Smoothing splines estimators in functional linear regression with errors-in-variablesComputational Statistics & Data Analysis, 2007
- Functional Data Analysis2005
- Semiparametric Regression2003