Abstract:Measurement error occurs when a covariate influencing a response variable is corrupted by noise. This can lead to misleading inference outcomes, particularly in problems where accurately estimating the relationship between covariates and response variables is crucial, such as causal effect estimation. Existing methods for dealing with measurement error often rely on strong assumptions such as knowledge of the error distribution or its variance and availability of replicated measurements of the covariates. We propose a Bayesian Nonparametric Learning framework that is robust to mismeasured covariates, does not require the preceding assumptions, and can incorporate prior beliefs about the error distribution. This approach gives rise to a general framework that is suitable for both Classical and Berkson error models via the appropriate specification of the prior centering measure of a Dirichlet Process (DP). Moreover, it offers flexibility in the choice of loss function depending on the type of regression model. We provide bounds on the generalization error based on the Maximum Mean Discrepancy (MMD) loss which allows for generalization to non-Gaussian distributed errors and nonlinear covariate-response relationships. We showcase the effectiveness of the proposed framework versus prior art in real-world problems containing either Berkson or Classical measurement errors.

Measurement errors models with exponential parametric multiplicative distortions

Parametric hypothesis tests for exponentiality under multiplicative distortion measurement errors data

Linear regression models with general distortion measurement errors

Parameter Estimation in Exponential Models by Linear and Nonlinear Fitting Methods.

A class of distortion measures generated from expectile and its estimation

The Quadratic Inference Functions in Measurement Error Model for Longitudinal Data

A revisit to Pearson correlation coefficient under multiplicative distortions

Checking normality of model errors under additive distortion measurement errors

Symmetry measures under additive distortion measurement errors

Measuring symmetry and asymmetry of multiplicative distortion measurement errors data

Non parametric regression models with additive distortions

Estimating Exponential Affine Models with Correlated Measurement Errors: Applications to Fixed Income and Commodities

Partial index additive models with additive distortion measurement errors

Jackknife empirical likelihood for the correlation coefficient with multiplicative distortion measurement errors

Robust Bayesian Inference for Berkson and Classical Measurement Error Models

Estimation on functional partially linear single index measurement error model

Model Averaging Estimation for Nonparametric Varying-Coefficient Models with Multiplicative Heteroscedasticity

Corruption of the Pearson correlation coefficient by measurement error and its estimation, bias, and correction under different error models

On Estimating the Nonparametric Multiplicative Error Models

Robust estimation in a nonlinear cointegration model

Estimators of the correlation coefficient in the bivariate exponential distribution