Adaptive Huber Regression

B-Tier
Journal: Journal of the American Statistical Association
Year: 2020
Volume: 115
Issue: 529
Pages: 254-265

Authors (3)

Qiang Sun (not in RePEc) Wen-Xin Zhou (not in RePEc) Jianqing Fan (Princeton University)

Score contribution per author:

0.670 = (α=2.01 / 3 authors) × 1.0x B-tier

α: calibrated so average coauthorship-adjusted count equals average raw count

Abstract

Big data can easily be contaminated by outliers or contain variables with heavy-tailed distributions, which makes many conventional methods inadequate. To address this challenge, we propose the adaptive Huber regression for robust estimation and inference. The key observation is that the robustification parameter should adapt to the sample size, dimension and moments for optimal tradeoff between bias and robustness. Our theoretical framework deals with heavy-tailed distributions with bounded (1+δ) th moment for any δ>0

Technical Details

RePEc Handle
repec:taf:jnlasa:v:115:y:2020:i:529:p:254-265
Journal Field
Econometrics
Author Count
3
Added to Database
2026-01-25