Open Access
September, 1984 Infinitesimal Robustness for Autoregressive Processes
H. Kunsch
Ann. Statist. 12(3): 843-863 (September, 1984). DOI: 10.1214/aos/1176346706

Abstract

We define the influence function and construct optimal robust estimators for autoregressive processes. We show that the asymptotic bias caused by small contaminations of the marginals can be written as the integral of a certain function with respect to the contamination. This function is called the influence function. It is unique only up to an equivalence relation, but there is a natural unique version which describes the limiting influence of an additional observation given the previous observations. Moreover, with this version the asymptotic variance at the true model can be expressed in a simple form. Optimal robust estimators minimize this asymptotic variance under a constraint on the influence function. As in the i.i.d case, they are found by truncating a multiple of the influence function of the maximum likelihood estimator.

Citation

Download Citation

H. Kunsch. "Infinitesimal Robustness for Autoregressive Processes." Ann. Statist. 12 (3) 843 - 863, September, 1984. https://doi.org/10.1214/aos/1176346706

Information

Published: September, 1984
First available in Project Euclid: 12 April 2007

zbMATH: 0587.62077
MathSciNet: MR751277
Digital Object Identifier: 10.1214/aos/1176346706

Subjects:
Primary: 62F35
Secondary: 62M10

Keywords: $M$-estimators , asymptotic properties , autoregressive processes , contaminations of time series , influence function , optimal robust estimators , Robustness for time series

Rights: Copyright © 1984 Institute of Mathematical Statistics

Vol.12 • No. 3 • September, 1984
Back to Top