# MEBoost: Variable Selection in the Presence of Measurement Error

## TL;DR

MEBoost is a new iterative method for variable selection in regression models that accounts for measurement error in covariates, outperforming existing methods especially as measurement error increases.

## Contribution

The paper introduces MEBoost, a novel algorithm that corrects for measurement error during variable selection in regression models, improving accuracy over existing methods.

## Key findings

- MEBoost reduces prediction error with increased measurement error.
- MEBoost outperforms CoCoLasso and naive Lasso in variable selection accuracy.
- Application to clinical trial data demonstrates practical utility.

## Abstract

We present a novel method for variable selection in regression models when covariates are measured with error. The iterative algorithm we propose, MEBoost, follows a path defined by estimating equations that correct for covariate measurement error. Via simulation, we evaluated our method and compare its performance to the recently-proposed Convex Conditioned Lasso (CoCoLasso) and to the "naive" Lasso which does not correct for measurement error. Increasing the degree of measurement error increased prediction error and decreased the probability of accurate covariate selection, but this loss of accuracy was least pronounced when using MEBoost. We illustrate the use of MEBoost in practice by analyzing data from the Box Lunch Study, a clinical trial in nutrition where several variables are based on self-report and hence measured with error.