Robust Estimation for ARMA models

NORA MULER; DANIEL PEÑA; VÍCTOR J. YOHAI

ANNALS OF STATISTICS, THE

Institute of Mathematical Statistics

Lugar: Philadelphia, PA, USA; Año: 2009 vol. 37 p. 816 - 840

0090-5364

This paper introduces a new class of robust estimates for ARMA models. They are M-estimates, but the residuals are computed so that the e¤ect of one outlier is is limited to the period where it occurs. These estimates are closely related to those based on a robust .lter, but they have two important advantages: they are consistent and the asymptotic theory is tractable. We perform a Monte Carlo where we show that these estimates compare favorable with respect to standard M-estimates and to estimates based on a diagnostic procedure.