Robust Generalized Empirical Likelihood for Heavy Tailed Autoregressions with Conditionally Heteroscedastic Errors

35 Pages Posted: 8 Feb 2013 Last revised: 18 Jan 2015

See all articles by Jonathan B. Hill

Jonathan B. Hill

University of North Carolina (UNC) at Chapel Hill – Department of Economics

Date Written: January 18, 2015

Abstract

We present a robust Generalized Empirical Likelihood estimator and confidence region for the parameters of an autoregression that may have a heavy tailed error, and the error may be conditionally heteroscedastic of unknown form. The estimator exploits two transformations for heavy tail robustness: a redescending transformation of the errors that robustifies against innovation outliers, and weighted least squares instruments that ensure robustness against heavy tailed regressors. Our estimator is consistent for the true parameter and asymptotically normal irrespective of heavy tails.

Keywords: Empirical Likelihood, autoregression, tail trimming, tail estimation

JEL Classification: C13, C22

Suggested Citation

Hill, Jonathan B., Robust Generalized Empirical Likelihood for Heavy Tailed Autoregressions with Conditionally Heteroscedastic Errors (January 18, 2015). Available at SSRN: https://ssrn.com/abstract=2213457 or http://dx.doi.org/10.2139/ssrn.2213457

Jonathan B. Hill (Contact Author)

University of North Carolina (UNC) at Chapel Hill – Department of Economics ( email )

102 Ridge Road
Chapel Hill, NC NC 27514
United States