Scalar-on-function regression with measurement error in the functional regressors
Abstract
We consider the problem of scalar-on-function regression. Most existing methods implicitly assume that the functional covariates are observed exactly, but in practice, they are often contaminated by measurement error. Our goal, therefore, is to deal with the problem of scalar-on-function regression when the regressor functions are observed with error. In this paper, we propose a simulation-extrapolation method to correct for the attenuation of estimated coefficient functions caused by the error. The method first estimates the error variance, establishes the relationship between a sequence of added error variance and the corresponding estimates of coefficient functions, and then extrapolates to the zero-error. We describe three methods to extrapolate the sequence of estimated coefficient functions. In a simulation study, we compare the performance of the simulation-extrapolation method with two pre-smoothing methods based on smoothing splines and functional principal component analysis. Next, we discuss the extension of the method in several directions, allowing for more complex noise covariance structures, multiple replications of functional predictors, generalized responses, and 2D and 3D functional predictors. Finally, we illustrate the methods by an application to diffusion tensor imaging data.
Related
- Nonparametric Regression with Measurement Error in Banach Spaces
- Subdata Selection for Measurement-Constrained Linear Regression under Model Misspecification
- Debiasing ML- or AI-Generated Regressors in Partially Linear Models
- Estimation and Inference for One-pass Streaming Function-on-Function Regression
- SPACR: Single-Pass Adaptive Training of Uncertainty-Aware Conformal Regressors
BibTeX
@misc{cai2026scalaronfunction,
title = {{Scalar-on-function regression with measurement error in the functional regressors}},
author = {Xiaochen Cai and Fenglin Xie and Todd Ogden},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.00867},
eprint = {2610.00867},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.00867},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Scalar-on-function regression with measurement error in the functional regressors
AU - Xiaochen Cai
AU - Fenglin Xie
AU - Todd Ogden
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.00867
UR - https://arxiv.org/abs/2610.00867
AB - We consider the problem of scalar-on-function regression. Most existing methods implicitly assume that the functional covariates are observed exactly, but in practice, they are often contaminated by measurement error. Our goal, therefore, is to deal with the problem of scalar-on-function regression when the regressor functions are observed with error. In this paper, we propose a simulation-extrapolation method to correct for the attenuation of estimated coefficient functions caused by the error. The method first estimates the error variance, establishes the relationship between a sequence of added error variance and the corresponding estimates of coefficient functions, and then extrapolates to the zero-error. We describe three methods to extrapolate the sequence of estimated coefficient functions. In a simulation study, we compare the performance of the simulation-extrapolation method with two pre-smoothing methods based on smoothing splines and functional principal component analysis. Next, we discuss the extension of the method in several directions, allowing for more complex noise covariance structures, multiple replications of functional predictors, generalized responses, and 2D and 3D functional predictors. Finally, we illustrate the methods by an application to diffusion tensor imaging data.
ER -