Robust Estimation for Arma Models Articles uri icon

authors

  • PEÑA GIL, DANIEL
  • YOHAI, VICTOR JAIME
  • MULER, NORA

publication date

  • April 2009

start page

  • 816

end page

  • 840

issue

  • 2

volume

  • 37

International Standard Serial Number (ISSN)

  • 0090-5364

Electronic International Standard Serial Number (EISSN)

  • 0003-4851

abstract

  • This paper introduces a new class of robust estimates for ARMA models. They are M-estimates, but the residuals are computed so the effect of one outlier is limited to the period where it occurs. These estimates are closely related to those based on a robust filter, 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 favorably with respect to standard M-estimates and to estimates based on a diagnostic procedure.