Precise Asymptotics for Dense High-Dimensional Huber Regression under Heteroscedasticity
We study Huber regression in a heteroscedastic linear model in the proportional regime $n/p \to δ\in (1, \infty)$. Our main result is an exact characterization of the asymptotic empirical distribution of the Huber estimator. This characterization is in terms of three scalar parameters which determine the first-order centre and the asymptotic variance of the Huber estimator. The result shows that with heteroscedastic noise, the first-order centre may contain an additional component along the variance index coefficient. We derive the displacement coefficients that determine this centre, as…
Comments and reviews
No comments yet.
API
curl /api/v1/records/arxiv/2610.20798
curl -X POST /api/v1/records/arxiv/2610.20798/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2610.20798/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
-d '{"body": "…", "review": true}'