Home > Research > Publications & Outputs > Model selection based on combined penalties for...

Electronic data

  • stepModelSelectPenalties_JBSresubmit

    Rights statement: This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Biopharmaceutical Statistics on 26/10/2017, available online: http://www.tandfonline.com/10.1080/10543406.2017.1378662

    Accepted author manuscript, 404 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Model selection based on combined penalties for biomarker identification

Research output: Contribution to journalJournal article

Published
Close
<mark>Journal publication date</mark>4/07/2018
<mark>Journal</mark>Journal of Biopharmaceutical Statistics
Issue number4
Volume28
Number of pages15
Pages (from-to)735-749
Publication statusPublished
Early online date26/10/17
Original languageEnglish

Abstract

The growing role of targeted medicine has led to an increased focus on the development of actionable biomarkers. Current penalized selection methods that are used to identify biomarker panels for classification in high-dimensional data, however, often result in highly complex panels that need careful pruning for practical use. In the framework of regularization methods, a penalty that is a weighted sum of the L1 and L0 norm has been proposed to account for the complexity of the resulting model. In practice, the limitation of this penalty is that the objective function is non-convex, non-smooth, the optimization is computationally intensive and the application to high-dimensional settings is challenging. In this paper, we propose a stepwise forward variable selection method which combines the L0 with L1 or L2 norms. The penalized likelihood criterion that is used in the stepwise selection procedure results in more parsimonious models, keeping only the most relevant features. Simulation results and a real application show that our approach exhibits a comparable performance with common selection methods with respect to the prediction performance while minimizing the number of variables in the selected model resulting in a more parsimonious model as desired.

Bibliographic note

This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Biopharmaceutical Statistics on 26/10/2017, available online: http://www.tandfonline.com/10.1080/10543406.2017.1378662