Global VAR Modelling

Macroeconomic policy analysis and risk management require taking account of the increasing interdependencies that exist across markets and countries. This invariably means that many different channels of transmission must be taken into consideration. These increased interdependencies could be due to common observed global shocks (such as changes in oil prices), they could arise as a result of global unobserved factors (such as the diffusion of technological progress), or could be due to specific national or sectoral shocks. It is also likely that even when all "common" factors are allowed for, there can be important residual interdependencies due to policy and trade spillover effects that remain to be explained. Therefore, a fairly detailed global framework is needed if we are to investigate the relative importance of such diverse sources of comovements in the world economy.

The GVAR (Global Vector AutoRegressive) methodology provides a general, yet practical, global modelling framework for the quantitative analysis of the relative importance of different shocks and channels of transmission mechanisms. It is a relatively novel approach to global macroeconomic modelling that combines time series, panel data, and factor analysis techniques to address a wide set of economic and financial issues ranging from policy analysis to risk management.

In particular, the GVAR comprises a compact model of the world economy designed to explicitly model the economic and financial interdependencies at national and international levels. It combines individual country/regional vector error-correcting models, where the domestic variables are related to corresponding foreign variables constructed exclusively to match the international trade, financial or other, desired patterns of the country under consideration. The individual country models are linked in a consistent manner so that the GVAR model is solved for the world as a whole. It can then be used to investigate the degree of regional interdependencies via impulse response analysis. The framework relies exclusively on observables, which typically include macroeconomic aggregates and financial variables, with the country-specific foreign variables serving as a proxy for common unobserved factors.

The key advantages of the GVAR modelling approach are that it:

• allows for interdependence at a variety of levels (national and international) in a transparent way that can be empirically evaluated

• allows for long-run relationships consistent with the theory and short-run relationships that are consistent with the data

• provides a coherent, theory-consistent solution to the curse of dimensionality in global modelling.

While the GVAR model is a suitable tool for policy analysis, its use is much broader. It has been used for analysing credit risk (Pesaran, Schuermann, Treutler, and Weiner (2006), for forecasting purposes (Pesaran, Schuermann and Smith, 2009), as well as for counterfactual analysis, such as to evaluate the UK entry into the Euro (Pesaran, Smith and Smith, 2007). For a broad cover of GVAR applications see the GVAR Handbook and Chudik and Pesaran (2014).

The model was originally introduced in Pesaran, Schuermann and Weiner (2004) where 11 country/region models were estimated over the period 1979Q1-1999Q4. The GVAR framework was subsequently extended in a number of ways in Dees, di Mauro Pesaran and Smith (2007, DdPS), in collaboration with the European Central Bank. The DdPS version of the GVAR model includes 26 countries/regions (covering 90% of world output), the euro area being treated as a single economy, and is estimated over the period 1979Q1-2003Q4.

For more on global modelling see Global and National Macroeconometric Modelling: A Long Run Structural Approach, by Tony Garrett, Kevin Lee, Hashem Pesaran and Yongcheol Shin, Oxford University Press, 2006. ISBN 0-19-929685-5.

To implement the GVAR modelling approach you can use the GVAR Toolbox, which is accompanied by a comprehensive user guide. Click here to learn more>