Impulse Response Tool Calculator

Estimate shock propagation with flexible periods, damping, and confidence ranges. Compare immediate and cumulative effects. Turn lagged inputs into useful response insights for decisions.

Calculator inputs

Use the form below to model how a one-time shock moves through a lagged statistical process. The form uses a responsive three-column layout on large screens, two columns on smaller screens, and one column on mobile.

Magnitude of the one-period external shock.
Number of future periods to evaluate.
Persistence from the previous response state.
Immediate propagation from the innovation term.
Additional decay or amplification across periods.
Reference level before the shock is applied.
Used to estimate confidence intervals.
Larger samples narrow approximate confidence bands.
Applies a z-score to each estimated period.
Useful for comparing different shock sizes.
Adds cumulative effect for easier interpretation.

Example data table

The following example shows a compact interpretation of a typical response sequence from a positive shock in a moderately persistent process.

Period Shock IR Weight Estimated Response Cumulative Effect
0 1.00 1.0000 1.0000 1.0000
1 0.00 0.8075 0.8075 1.8075
2 0.00 0.4986 0.4986 2.3061
3 0.00 0.3079 0.3079 2.6140
4 0.00 0.1901 0.1901 2.8041

Formula used

Impulse response weights
ψ0 = 1
ψh = φh-1(φ + θ)δh, for h ≥ 1
Period response
IRFh = Shock × ψh
Approximate confidence interval
SEh = (σ / √n) × √(1 + h)
Lowerh = IRFh − z × SEh
Upperh = IRFh + z × SEh
Cumulative response
CumIRFh = Σ IRFj, from j = 0 to h
Long-run multiplier
LRM = 1 + [δ(φ + θ) / (1 − φδ)], when the denominator is not zero

This tool uses a practical ARMA-style response framework with an additional damping factor. It is suited for exploratory statistics, teaching, scenario testing, and fast comparisons across different parameter settings.

How to use this calculator

  1. Enter the one-time shock size you want to simulate.
  2. Select the number of future periods to trace.
  3. Provide AR and MA coefficients to control persistence and immediate spillover.
  4. Set a damping factor to reflect extra decay or amplification.
  5. Add baseline level, innovation standard deviation, and sample size.
  6. Choose a confidence level for uncertainty bands.
  7. Enable normalization when you want responses per shock unit.
  8. Submit the form to see results above the calculator, including the graph and downloadable table.

Frequently asked questions

1. What does an impulse response measure?

It measures how a one-time shock affects future periods in a dynamic process. The response can show persistence, reversal, damping, or amplification over time.

2. Why are AR and MA coefficients both included?

The AR coefficient controls persistence from prior states. The MA coefficient captures immediate propagation from the innovation. Together they give a richer short-run and medium-run response pattern.

3. What does the damping factor do?

The damping factor adds extra decay or growth across future periods. Values below one dampen the response. Values above one amplify it and may create unstable trajectories.

4. What does normalization change?

Normalization divides response values by the shock size, making outputs comparable across scenarios. It is useful when you want a response per unit shock instead of a raw effect.

5. How should I read the confidence bounds?

The bounds provide an approximate uncertainty range around each period response. Wider bands indicate greater uncertainty, while narrower bands suggest more precise period estimates.

6. What does the cumulative line mean?

The cumulative line adds all past period responses through the selected horizon. It helps you understand the total accumulated influence of the original shock.

7. When is a process considered stable here?

This tool flags the system as stable when |φ × damping| is below one. That suggests the effect should fade instead of exploding over time.

8. Is this suitable for formal model inference?

It is best for exploration, teaching, and quick scenario analysis. Formal inference should rely on a fully estimated model, validated assumptions, and robust diagnostic testing.

Related Calculators

Moving Average CalculatorSeasonal Decomposition ToolTrend Analysis CalculatorTime Series Forecast ToolPartial Autocorrelation ToolStationarity Test ToolADF Test CalculatorKPSS Test CalculatorHolt Linear TrendHolt Winters Tool

Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.