Article

Practical methods for modelling weak VARMA processes: identification, estimation and specification with a macroeconomic application

Department of Economics, McGill University, H3A 2T7, Montréal, Québec, Canada
11/2008;

ABSTRACT program on Mathematics of Information Technology and Complex Systems (MITACS)], the Canada Council for the Arts (Killam Fellowship), the CIREQ, the CIRANO, and the Fonds FCAR (Government of Québec). William Dow Professor of Economics, McGill University, Centre interuniversitaire de recherche en analyse des organisations (CIRANO), and Centre interuniversitaire de recherche en économie quantitative (CIREQ). Mailing address: ABSTRACT In this paper, we develop practical methods for modelling weak VARMA processes. In a first part, we propose new identified VARMA representations, the diagonal MA equation form and the final MA equation form, where the MA operator is diagonal and scalar respectively. Both of these representations have the important feature that they constitute relatively simple modifications of a VAR model (in contrast with the echelon representation). In a second part, we study the problem of estimating VARMA models by relatively simple methods which only require linear regressions. We consider a generalization of the regression-based estimation method proposed by Hannan and Rissanen (1982). The asymptotic properties of the estimator are derived under weak hypotheses on the innovations (uncorrelated and strong mixing) so as to broaden the class of models to which it can be applied. In a third part, we present a modified information criterion which gives consistent estimates of the orders under the proposed representations. To demonstrate the importance of using VARMA models to study multivariate time series we compare the impulse-response functions and the out-of-sample forecasts generated by VARMA and VAR models.

Download full-text

Full-text

Available from: Jean-Marie Dufour, Dec 16, 2013
1 Follower
 · 
168 Views
  • Source
    • "The works on the multivariate version of the models selection are generally performed under the assumption that the errors t are independent (see also Reinsel, 1997, Lü tkepohl, 2005). Under weak assumptions on the innovation process, a notable exception is Dufour and Pelletier (2005) who have proposed a modified information criterion, which is a generalization of the information criterion proposed by Hannan and Rissanen (1982). For the statistical inference of weak VARMA models, important advances have been obtained by Dufour and Pelletier (2005) who study the asymptotic properties of a generalization of the regression-based estimation method proposed by Hannan and Rissanen (1982) under weak assumptions on the innovation process, Francq and Raı¨ssi (2007) and Boubacar Mainassara (2011) who study respectively portmanteau tests for weak VAR and structural VARMA models. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This article considers the problem of order selection of the vector autoregressive moving‐average (VARMA) models under the assumption that the errors are uncorrelated but not necessarily independent. These models are called weak VARMA by opposition to the standard VARMA models, also called strong VARMA models, in which the error terms are supposed to be i.i.d. We relax the standard independence assumption to extend the range of application of the VARMA models, allowing us to treat linear representations of general nonlinear processes. We propose a modified version of the Akaike information criterion for identifying the orders of weak VARMA models.
    Journal of Time Series Analysis 01/2012; 33(1). DOI:10.1111/j.1467-9892.2011.00746.x · 0.81 Impact Factor
  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The concept of causality introduced by Wiener (1956) and Granger (1969) is defined in terms of predictability one period ahead. This concept can be generalized by considering causality at a given horizon h, and causality up to any given horizon h [Dufour and Renault (1998)]. This generalization is motivated by the fact that, in the presence of an auxiliary variable vector Z, it is possible that a variable Y does not cause variable X at horizon 1, but causes it at horizon h > 1. In this case, there is an indirect causality transmitted by Z. Another related problem consists in measuring the importance of causality between two variables. Existing causality measures have been defined only for the horizon 1 and fail to capture indirect causal effects. This paper proposes a generalization of such measures for any horizon h. We propose nonparametric and parametric measures of unidirectional and instantaneous causality at any horizon h. Parametric measures are defined in the context of autoregressive processes of unknown order and expressed in terms of impulse response coefficients. On noting that causality measures typically involve complex functions of model parameters in VAR and VARMA models, we propose a simple method to evaluate these measures which is based on the simulation of a large sample from the process of interest. We also describe asymptotically valid nonparametric confidence intervals, using a bootstrap technique. Finally, the proposed measures are applied to study causality relations at different horizons between macroeconomic, monetary and financial variables in the U.S. These results show that there is a strong effect of nonborrowed reserves on federal funds rate one month ahead, the effect of real gross domestic product on federal funds rate is economically important for the first three months, the effect of federal funds rate on gross domestic product deflator is economically weak one month ahead, and finally federal fundsrate causes the real gross domestic product until 16 months.
Show more