SIMEXBoost: An R package for Analysis of High-Dimensional Error-Prone Data Based on Boosting Method
Li‐Pang Chen, Bangxu Qiu · The R Journal · 2024
Boosting is a powerful statistical learning method. Its key feature is the ability to derive a strong learner from simple yet weak learners by iteratively updating the learning results. Moreover, boosting algorithms have been employed to do variable selection and estimation for regression models. However, measurement error usually appears in covariates. Ignoring measurement error can lead to biased estimates and wrong inferences. To the best of our knowledge, few packages have been developed to address measurement error and variable selection simultaneously by using boosting algorithms. In this paper, we introduce an R package [SIMEXBoost](https://CRAN.R-project.org/package=SIMEXBoost), which covers some widely used regression models and applies the simulation and extrapolation method to deal with measurement error effects. Moreover, the package [SIMEXBoost](https://CRAN.R-project.org/package=SIMEXBoost) enables us to do variable selection and estimation for high-dimensional data under various regression models. To assess the performance and illustrate the features of the package, we conduct numerical studies.