Quantile VAR
Also known as: Quantile-based impulse response
Quantile VAR estimates impulse responses of multivariate systems conditional on different quantiles of the distribution, revealing how shocks propagate heterogeneously across the conditional distribution. Introduced by Koenker and Xiao (2006) and applied to risk measurement by White et al. (2015), it reveals tail behavior and contagion effects invisible to mean-based VAR analysis. This is essential for risk management and understanding how crises propagate differently than normal times.
Key highlights
- Captures tail behavior and extreme-event dynamics
- Reveals heterogeneous shock transmission across regimes
- Quantile-specific impulse responses enable risk-focused analysis
- No parametric distribution assumptions; fully nonparametric
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use Quantile VAR when risk management and tail behavior are central (banking, insurance, options), when heterogeneity across regimes matters (crises vs normal times transmit differently), or when you suspect nonlinear shock transmission. Essential for Value-at-Risk analysis and stress testing.
Strengths & limitations
- Captures tail behavior and extreme-event dynamics
- Reveals heterogeneous shock transmission across regimes
- Quantile-specific impulse responses enable risk-focused analysis
- No parametric distribution assumptions; fully nonparametric
- Estimation is computationally intensive; standard errors difficult to compute reliably
- Cross-quantile dependence often ignored, potentially biasing standard errors
- Interpretation of quantile-specific coefficients in dynamic systems is subtle
- Results can be sensitive to quantile specification and lag length
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
Which quantiles should I estimate?
For risk (VaR), focus on tails: 1st, 5th, 25th, 50th, 75th, 95th, 99th percentiles. For comprehensive analysis, estimate deciles. Balance detail against precision loss at extreme quantiles.
How do I handle cross-quantile dependence in inference?
Bootstrap methods respecting the full dependence structure are most reliable. Analytical standard errors assuming independence are often too tight. Use quantile-regression software's built-in robust inference when available.
Can I identify shocks in Quantile VAR?
Yes, using sign restrictions or zero restrictions. Quantile-specific identification is possible but complex; typically assume common identification structure across quantiles for simplicity.
How do Quantile-VAR and Quantile-ARDL impulse responses compare?
Quantile-VAR includes lagged dependent variables; Quantile-ARDL is more flexible but parsimonious. Quantile-VAR provides fuller dynamic feedback; Quantile-ARDL is better for short-horizon policy analysis.
Sources
- 1.Koenker, R., & Xiao, Z. (2006). Quantile autoregression. Journal of the American Statistical Association, 101(475), 980-990.
- 2.White, H., Kim, T. H., & Manganelli, S. (2015). VAR for VaR: Measuring tail dependence using multivariate regression quantiles. Journal of Econometrics, 187(1), 169-188.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 3). Quantile VAR. ScholarGate. https://scholargate.app/econometrics/quantile-var