SIMEX-Trans: High-Dimensional Transfer Learning With Covariates Subject to Measurement Error.

Authors

Kaida Cai, Zhengyan Wang, Hanning Chen, Hanwen Liu

Published in

Statistics in medicine. Volume 45. Issue 25-27.

Abstract

Gene expression studies typically involve small sample sizes, high-dimensional covariates, and measurement error, posing significant challenges for accurate statistical modeling. We propose simulation-extrapolation (SIMEX)-based high-dimensional transfer learning methods (SIMEX-Trans) that (i) correct measurement error and (ii) leverage information from informative auxiliary studies. We consider settings where the informative auxiliary studies are either known or unknown, and develop Lasso- and SCAD-penalized estimators. When the informative studies are known, we establish support recovery and asymptotic normality for the SIMEX-based SCAD-penalized estimator. Simulation results show that SIMEX-Trans outperforms competing methods in estimation accuracy and achieves reliable variable selection performance. In a GTEx brain-tissue analysis of cross-tissue prediction of hypothalamus JAM2 expression with limited target samples and heterogeneous auxiliary tissues, the proposed SIMEX-Trans-Selector approach achieves the lowest cross-validated prediction error and remains stable under sensitivity analysis.

PMID:
43016007
Bibliographic data and abstract were imported from PubMed on 12 Nov 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.