HRLR regression. Issue 1 (1st January 2021)

• Record Type: Journal Article
• Authors: Pelawa Watagoda, Lasanthi C. R.; Arnholt, Alan T.; Arachchige Don, Hasthika S. Rupasinghe
• Is Part Of: RMS: Research in mathematics & statistics. Volume 8:Issue 1(2021)
• Publication Date: 2021-01-01
• Subjects: Lasso -- relaxed lasso -- variable section -- model selection -- ridge regression -- elastic net -- p≫n problem; Mathematics -- Periodicals; Statistics -- Periodicals; 510
• DOI: 10.1080/27658449.2021.1921904
• Languages: English
• ISSNs: 2765-8449

Abstract

ABSTRACT: Selecting the most important predictor variables and achieving high prediction accuracy are the two main goals in Statistical Learning. In this work, we propose a new regularization method, HRLR (a H ybrid of R elaxed L asso and R idge Regression ), where we use properties of both relaxed lasso and ridge regression. We also demonstrate the effectiveness of our method and the accuracy of the algorithm on simulated as well as real-world data. Simulation results suggest that the HRLR regression outperforms the existing well-known methods including lasso and Elastic net in many scenarios described, especially when the error distribution is non-normal. The proposed algorithm is implemented in R and links are provided in this paper along with the R functions used for the simulation.