Saved in:
Bibliographic Details
Main Author: Du, Kasy
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.07655
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917935515172864
author Du, Kasy
author_facet Du, Kasy
contents This paper compares convex and non-convex penalized likelihood methods in high-dimensional statistical modeling, focusing on their strengths and limitations. Convex penalties, like LASSO, offer computational efficiency and strong theoretical guarantees but often introduce bias in parameter estimation. Non-convex penalties, such as SCAD and MCP, reduce bias and achieve oracle properties but pose optimization challenges due to non-convexity. The paper highlights key differences in bias-variance trade-offs, computational complexity, and robustness, offering practical guidance for method selection. It concludes that the choice depends on the problem context, balancing accuracy
format Preprint
id arxiv_https___arxiv_org_abs_2502_07655
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Short Note of Comparison between Convex and Non-convex Penalized Likelihood
Du, Kasy
Methodology
This paper compares convex and non-convex penalized likelihood methods in high-dimensional statistical modeling, focusing on their strengths and limitations. Convex penalties, like LASSO, offer computational efficiency and strong theoretical guarantees but often introduce bias in parameter estimation. Non-convex penalties, such as SCAD and MCP, reduce bias and achieve oracle properties but pose optimization challenges due to non-convexity. The paper highlights key differences in bias-variance trade-offs, computational complexity, and robustness, offering practical guidance for method selection. It concludes that the choice depends on the problem context, balancing accuracy
title A Short Note of Comparison between Convex and Non-convex Penalized Likelihood
topic Methodology
url https://arxiv.org/abs/2502.07655