Skip to main content

Cookies on Macro by Mark.We use essential cookies for sign-in, locale, and security. With your permission, we also use optional cookies for preferences, for analytics, and for advertising measurement with the Meta (Facebook) Pixel. Read our cookie policy.

Macro by MarkMacro by Mark
  • Home
Loading…
Loading…
← Models
MacroConceptsPolicyModels

Panel VAR Model

Vector autoregression estimated across a panel of countries or regions to study common dynamics and shock propagation.

OverviewGraphMathReferences

Graph workspace

Panel impulse response functions with cross-unit dispersion bands

Graph workspace

The defining output of a panel VAR is the impulse response function, but unlike a single-unit VAR, the panel version reveals cross-unit heterogeneity. Plotting the pooled/mean IRF with a dispersion fan showing the range across units communicates both the average dynamic and the degree of heterogeneity - the two quantities practitioners care about most.

Chart axis metadata
xHorizon (quarters after shock)quarters · 0 - 20yResponse of output (percentage points)percentage points · -2.0 to 1.0

Shaded ranges come from assumed parameters in a simulated example. They are not estimated from observed data.

Mean response across units static chart preview showing Mean response across units
Upper quartile across units static chart preview showing Upper quartile across units
Lower quartile across units static chart preview showing Lower quartile across units
xHorizon (quarters after shock)quarters · 0 - 20yResponse of output (percentage points)percentage points · -2.0 to 1.0

Current results

Mean peak response

Explain: Mean peak response
min⁡hIRF‾h\min_h \overline{\text{IRF}}_hminh​IRFh​

Largest average output decline across all units following the shock: the lowest point of the mean response over the 24 plotted quarters, reached after the impact jump as the cross channel (a12)(a_{12})(a12​) pulls output below baseline. Its quarter, h∗h^*h∗, is where the next readouts are measured. A positive value means the average never falls below baseline.

-0.01 pp

percentage points

Cross-unit IQR at peak

Explain: Cross-unit IQR at peak
Q75(IRFi,h∗)−Q25(IRFi,h∗)Q_{75}(\text{IRF}_{i,h^*}) - Q_{25}(\text{IRF}_{i,h^*})Q75​(IRFi,h∗​)−Q25​(IRFi,h∗​)

Interquartile spread of unit-specific responses at the horizon of peak mean response. Measures transmission heterogeneity.

0.04 pp

percentage points

Mean half-life

Explain: Mean half-life
min⁡{h:∣IRF‾h∣<0.5∣IRF‾0∣}\min\{h : |\overline{\text{IRF}}_h| < 0.5 |\overline{\text{IRF}}_0|\}min{h:∣IRFh​∣<0.5∣IRF0​∣}

Number of quarters until the average response decays to half its impact value, counted in whole quarters on the mean path.

2.00

quarters

Share of units with sign reversal

Explain: Share of units with sign reversal
N−1∑i1[sign⁡(IRFi,h∗)=−sign⁡(IRF‾h∗)]N^{-1} \sum_i \mathbf{1}[\operatorname{sign}(\text{IRF}_{i,h^*}) = -\operatorname{sign}(\overline{\text{IRF}}_{h^*})]N−1∑i​1[sign(IRFi,h∗​)=−sign(IRFh∗​)]

Fraction of units whose output response has the opposite sign from the mean at the peak horizon. Flags genuinely heterogeneous transmission.

0.3333

proportion

Controls

Shocks fade slowly
Explain: How long output shocks last

Output persistence: 0.7

How much of last period's output carries forward; higher values make output shocks last longer.

Raising a11a_{11}a11​ stretches the IRF horizontally - the mean response decays more slowly and the peak shifts rightward.

Mild drag on output
Explain: How much tightening cuts output

Monetary transmission: -0.15

How strongly a rate or inflation move feeds into output; more negative means tightening bites harder.

More negative a12a_{12}a12​ deepens the trough output falls into after the impact jump, as the rate rise feeds through.

Units differ a little
Explain: How much the units differ

Cross-unit heterogeneity: 0.08

How far each unit's response strays from the average; zero means every unit behaves identically.

Increasing τ\tauτ widens the dispersion bands. At zero, bands collapse to a single line.

12
Explain: How many units

How many units are in the sample; more units give smoother bands and a more precise average.

More units smooth the dispersion bands and stabilize the mean IRF line. Fewer units make bands jagged.

1
Explain: Size of the initial shock

How large the starting shock is, measured in standard deviations.

Scales the entire IRF vertically - larger shock, deeper trough, wider bands in absolute terms.

Keep going

Read next

Continue with a related concept, model, or example.

Theory model

IS-LM Model

a well-connected theoretical model two steps from here

New angle: theoretical model →
Macro by Mark
Help
Privacy PolicyTerms of UseEthics and Compliance

© 2026 Mark Jayson Nation LLC