Abstract:Contamination of covariates by measurement error is a classical problem in multivariate regression, where it is well known that failing to account for this contamination can result in substantial bias in the parameter estimators. The nature and degree of this effect on statistical inference is also understood to crucially depend on the specific distributional properties of the measurement error in question. When dealing with functional covariates, measurement error has thus far been modelled as additive white noise over the observation grid. Such a setting implicitly assumes that the error arises purely at the discrete sampling stage, otherwise the model can only be viewed in a weak (stochastic differential equation) sense, white noise not being a second-order process. Departing from this simple distributional setting can have serious consequences for inference, similar to the multivariate case, and current methodology will break down. In this paper, we consider the case when the additive measurement error is allowed to be a valid stochastic process. We propose a novel estimator of the slope parameter in a functional linear model, for scalar as well as functional responses, in the presence of this general measurement error specification. The proposed estimator is inspired by the multivariate regression calibration approach, but hinges on recent advances on matrix completion methods for functional data in order to handle the nontrivial (and unknown) error covariance structure. The asymptotic properties of the proposed estimators are derived. We probe the performance of the proposed estimator of slope using simulations and observe that it substantially improves upon the spectral truncation estimator based on the erroneous observations, i.e., ignoring measurement error. We also investigate the behaviour of the estimators on a real dataset on hip and knee angle curves during a gait cycle.

Functional Regression on the Manifold with Contamination

Functional Regression on Manifold with Contamination

Functional Regression with Unknown Manifold Structures

Functional Linear Regression: Dependence and Error Contamination

Regression with Functional Errors-in-Predictors: A Generalized Method-of-Moments Approach

Conformal inference for regression on Riemannian Manifolds

Partially Functional Linear Regression in High Dimensions

Functional Linear Regression with Mixed Predictors

Statistical Regression Analysis of Functional and Shape Data

Functional Regression with Mode-Sparsity Constraint

Supervised Principal Component Regression for Functional Responses with High Dimensional Predictors

Nonparametric Classification on Low Dimensional Manifolds using Overparameterized Convolutional Residual Networks

Functional Linear Regression for Discretely Observed Data: from Ideal to Reality

Local polynomial regression on unknown manifolds

High dimensional test for functional covariates

Deep Nonparametric Regression on Approximate Manifolds: Non-Asymptotic Error Bounds with Polynomial Prefactors

Robust Functional Regression with Discretely Sampled Predictors

Regression with genuinely functional errors-in-covariates

Single and multiple index functional regression models with nonparametric link

Theory of functional principal component analysis for discretely observed data

Diffusion-based Semi-supervised Spectral Algorithm for Regression on Manifolds