Global VAR
Global Vector Autoregression · Also known as: GVAR, Multi-country VAR
Global VAR (GVAR) is a large-scale macroeconomic modeling framework linking multiple countries (or regions) via trade and financial channels, allowing shocks in one country to propagate through the global system. Introduced by Pesaran et al. (2004), it solves the curse of dimensionality in international VAR models by estimating country-specific VARs conditional on foreign variables, then solving a system linking all countries. This approach is invaluable for analyzing global spillovers and international policy coordination.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use GVAR when studying international spillovers, global monetary transmission, or trade-flow responses to shocks. It is essential for central banks, international organizations, and researchers analyzing globally integrated economies. Requires data for multiple countries over sufficient time periods.
Strengths & limitations
- Handles large number of countries parsimoniously
- Naturally captures bilateral trade and financial linkages
- Generates global impulse responses showing international transmission
- Flexible for various identification schemes and extensions
- Requires synchronized data across many countries; missing observations complicate the system
- Weights (trade shares, financial flows) must be specified; results can be sensitive
- Interpretation of country-specific responses within the global system is complex
- Estimation of uncertainty (confidence bands) is computationally intensive
Frequently asked
How should I weight foreign variables?
Bilateral trade shares are most common; use lagged shares to avoid simultaneity. Financial linkages (bank lending, FDI) are increasingly important; combine trade and financial weights for comprehensive modeling.
What if a country is not in the system?
Small open economies can be dropped; their omission has minimal impact if they are not major shocks sources. Alternatively, group them into regions and include as one unit.
How do I identify shocks in GVAR?
Use structural identification (Cholesky ordering, sign restrictions, or theory-based). Global shocks versus idiosyncratic require careful identification strategy.
Can GVAR handle cointegration across countries?
Yes. Estimate error-correction versions of country VARs (VECM). Cointegration relationships can be specified per country or globally.
Sources
- Pesaran, M. H., Schuermann, T., & Weiner, S. M. (2004). Modeling regional interdependencies using a global error-correcting macroeconometric model. Journal of Business and Economic Statistics, 22(2), 129-162. DOI: 10.1198/073500104000000019 ↗
- Chudik, A., & Pesaran, M. H. (2016). Theory and practice of GVAR modelling. Journal of Economic Surveys, 30(2), 165-197. link ↗
How to cite this page
ScholarGate. (2026, June 3). Global Vector Autoregression. ScholarGate. https://scholargate.app/en/econometrics/global-var
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Panel VARXEconometrics↔ compare
- Threshold Panel VAREconometrics↔ compare
- TVP-FAVAREconometrics↔ compare