Mixed causal–noncausal autoregressions with exogenous regressors

Alain Hecq, Joao Victor Issler, Sean Telg*

*Corresponding author for this work

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Mixed causal–noncausal autoregressive (MAR) models have been proposed to model time series exhibiting nonlinear dynamics. Possible exogenous regressors are typically substituted into the error term to maintain the MAR structure of the dependent variable. We introduce a representation including these covariates called MARX to study their direct impact. The asymptotic distribution of the MARX parameters is derived for a class of non-Gaussian densities. For a Student (Formula presented.) likelihood, closed-form standard errors are provided. By simulations, we evaluate the MARX model selection procedure using information criteria. We examine the influence of the exchange rate and industrial production index on commodity prices.

Original languageEnglish
Pages (from-to)328-343
Number of pages16
JournalJournal of Applied Econometrics
Volume35
Issue number3
Early online date20 Jan 2020
DOIs
Publication statusPublished - 1 Apr 2020

Fingerprint Dive into the research topics of 'Mixed causal–noncausal autoregressions with exogenous regressors'. Together they form a unique fingerprint.

Cite this