Asymptotic Theory for Iterated One-Step Huber-Skip Estimators

Johansen, Soren; Nielsen, Bent

doi:10.2139/ssrn.1967408

Download This Paper

Open PDF in Browser

Add Paper to My Library

Asymptotic Theory for Iterated One-Step Huber-Skip Estimators

University of Copenhagen Economics Discussion Paper No. 11-29

17 Pages Posted: 3 Dec 2011

See all articles by Soren Johansen

Soren Johansen

University of Copenhagen - Department of Economics; Aarhus University - CREATES

Bent Nielsen

University of Oxford - Nuffield Department of Medicine

Date Written: December 2, 2011

Abstract

Iterated one-step Huber-skip M-estimators are considered for regression problems. Each one-step estimator is a re-weighted least squares estimators with zero/one weights determined by the initial estimator and the data. The asymptotic theory is given for iteration of such estimators using a tightness argument. The results apply to stationary as well as non-stationary regression problems.

Keywords: Huber-skip, iteration, one-step M-estimators, unit roots

JEL Classification: C32

Suggested Citation: Suggested Citation

Johansen, Soren and Nielsen, Bent, Asymptotic Theory for Iterated One-Step Huber-Skip Estimators (December 2, 2011). University of Copenhagen Economics Discussion Paper No. 11-29, Available at SSRN: https://ssrn.com/abstract=1967408 or http://dx.doi.org/10.2139/ssrn.1967408