Background
The Vector Autoregressive (VAR) model is a commonly used statistical model in econometrics that captures the linear interdependencies among multiple time series data. Introduced in the late 20th century, VAR models differ from traditional autoregressive models by allowing for more than one evolving variable to be influenced by its own past values, and the past values of other variables in the system.
Historical Context
The development of the VAR model is attributed to Christopher A. Sims in 1980. Sims introduced VAR as an alternative to the large-scale structural econometric models, which faced criticism for their rigid structure and assumptions. The VAR model offered a more flexible, data-driven approach, thus becoming a staple for econometric analysis, especially in macroeconomics.
Definitions and Concepts
A Vector Autoregressive (VAR) model generalizes the standard univariate autoregressive (AR) model to capture the relationships between multiple time series. Mathematically, a VAR model of order \(p\) (VAR(\(p\))) can be expressed as:
\[ y_t = c + A_1 y_{t-1} + A_2 y_{t-2} + … + A_p y_{t-p} + e_t \]
Where:
- \( y_t \) is a vector of endogenous variables at time \( t \).
- \( c \) is a vector of constants (intercepts).
- \( A_i \) are matrices of coefficients for \( i = 1, 2, …, p \).
- \( e_t \) is a vector of error terms.
Major Analytical Frameworks
The application of the VAR model spans across various economic schools of thought, each leveraging the model according to their analytical frameworks.
Classical Economics
Classical economists typically rely less on empirical model structures like VAR, focusing more on theoretical constructs such as supply and demand equilibrium.
Neoclassical Economics
Neoclassical economists may use VAR models to quantify the response of various economic variables to policy changes or external shocks, largely supporting their model-based equilibrium analysis.
Keynesian Economics
Keynesian economists could use VAR models to analyze short-run dynamic interactions and forecasting, particularly in assessing interventions aimed at stabilizing macroeconomic fluctuations.
Marxian Economics
VAR models are used less frequently in Marxian economics, which focuses more on the capitalist structures and broader socio-economic changes rather than short-term quantitative forecasts.
Institutional Economics
For institutional economists, VAR models may aid in understanding the statistical relationships governed by institutional constraints and policy regimes.
Behavioral Economics
Behavioral economists might use VAR models to help capture how past behavior and psychological factors impact multiple economic variables over time.
Post-Keynesian Economics
Post-Keynesian use of VAR models could involve studying non-linear dynamics and how past values of economic variables affect future outcomes in a non-deterministic manner.
Austrian Economics
Austrian economists typically critique the heavy reliance on statistical models like VAR, advocating for a more qualitative and deductive approach to economics.
Development Economics
Development economists may utilize VAR models to examine the impact and interplay of developmental policies and economic variables over time in emerging markets.
Monetarism
Monetarists leverage VAR models to quantify the impact of changes in the money supply on macroeconomic variables such as inflation, output, and employment, often focusing on policy analysis.
Comparative Analysis
The introduction and integration of VAR models have created robust avenues for econometric analysis, allowing for richer interpretation and exploration compared to univariate models. Traditional models often impose arbitrary restrictions that can be statistically tested and potentially overruled by the flexibility inherent in the VAR structure.
Case Studies
Numerous case studies demonstrate the efficacy of VAR models in real-world analysis. For instance, the analysis of monetary policy impacts demonstrated via the VAR model by Sims (1980), and subsequent developments such as structural VAR (SVAR) extensions, show their applicability in capturing complex dynamics.
Suggested Books for Further Studies
- “Time Series Analysis” by James D. Hamilton
- “Applied Econometric Time Series” by Walter Enders
- “The Structural Econometric Time Series Analysis Approach” by Arnold Zellner and Franz C. Palm
Related Terms with Definitions
Autoregressive Model (AR)
A type of statistical model that describes a time series’ future values based on its historical values, specifically focusing on capturing relationships within a single variable series over different time intervals.
Structural VAR (SVAR)
A variant of the VAR model that incorporates a structural identification scheme, allowing for the imposition of economic theories and policy-related constraints into the dynamic relationship model.
Cointegration
A statistical property of time series variables where two or more non-stationary series