Premium
Mixed causal–noncausal autoregressions with exogenous regressors
Author(s) -
Hecq Alain,
Issler Joao Victor,
Telg Sean
Publication year - 2020
Publication title -
journal of applied econometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.878
H-Index - 99
eISSN - 1099-1255
pISSN - 0883-7252
DOI - 10.1002/jae.2751
Subject(s) - econometrics , autoregressive model , covariate , representation (politics) , commodity , class (philosophy) , gaussian , economics , variable (mathematics) , mathematics , nonlinear system , model selection , lag , index (typography) , term (time) , statistics , computer science , mathematical analysis , physics , quantum mechanics , artificial intelligence , politics , political science , computer network , world wide web , market economy , law
Summary 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 t 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.