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 with dependent error and missing observation in reproducing kernel Hilbert spaces

Functional Linear Regression with Mixed Predictors

Functional Adaptive Huber Linear Regression

Strong optimality of kernel functional regression in norms with partial response variables and applications

Double robust estimation of functional outcomes with data missing at random

Functional Linear Regression: Dependence and Error Contamination

Asymptotic Distributions of Nonparametric Regression Estimators for Longitudinal or Functional Data

Statistical inference for function-on-function linear regression

Structured functional additive regression in reproducing kernel Hilbert spaces.

Estimation in functional partially linear spatial autoregressive model

Robust estimation of the generalised partial linear model with missing covariates

Robust Functional Regression with Discretely Sampled Predictors

Regression estimation for continuous-time functional data processes with missing at random response

On the functional regression model and its finite-dimensional approximations

A new RKHS-based global testing for functional linear model

Inference for generalized partial functional linear regression

Functional partially linear single index models with measurement error in both functional and real covariates

Partially linear functional quantile regression in a reproducing kernel Hilbert space

A Bayesian approach to functional regression: theory and computation

Regression with genuinely functional errors-in-covariates

Composite quantile estimation in partial functional linear regression model with dependent errors