Quantile Regression with Generalized Doubly Regularized framework

• 발행기관: 호서대학교 기초과학연구소
• 저자명: Hosik Choi
• 간행물 정보: 『기초과학연구 논문집』제16권 제1호, 115~123쪽, 전체 9쪽
• 주제분류: 공학 > 공학일반
• 파일형태: PDF
• 발행일자: 2008.12.30

영문 초록

On highly correlated data, it is known that the doubly regularized quantile regression with Lx and L2 performs better than other regularized quantile regression methods. In this paper, we consider the general framework of quantile regression with : doubly regularized penalties Lx and L2. The proposed method includes the doubly regularization, incentive regularization and fused lasso regularization. From simulation analysis, we investigate its performance.

목차

Ⅰ. Introduction

Ⅱ. Methdologies

Ⅲ. Numerical Study

Ⅳ. Discussion

Ⅴ. Reference

키워드

Quantile regression doubly regularization variable selection