Skip to content

random_inertia_sv

Jump to bottom
Jaromír Beneš edited this page Mar 20, 2026 · 1 revision

Stochastic Volatility with Random Inertia (RandomInertiaSV)

Overview

The Random Inertia Stochastic Volatility VAR incorporates time-varying persistence in the VAR dynamics alongside stochastic volatility. This estimator allows both the volatility and the persistence (inertia) of shocks to vary over time, providing a flexible framework for modeling changing economic dynamics where both the magnitude and persistence of economic relationships evolve.

Methodology

The RandomInertiaSV specification features:

  • Constant VAR coefficients with Minnesota-type priors
  • Standard stochastic volatility for error terms
  • Time-varying inertia parameters that affect shock persistence
  • Random walk evolution for inertia parameters
  • Combined volatility and persistence modeling

This approach is particularly useful when the persistence of economic shocks varies over time, such as when monetary policy effectiveness or fiscal multipliers change across different periods.

Parameters

Minnesota-Style Prior Parameters

  • Autoregression: Prior on first-order autoregression

    • Range: [0, 1]
    • Default: 0.8
    • Controls the belief in random walk behavior for own lags

  • Lambda1: Overall tightness of priors

    • Range: (0, ∞)
    • Default: 0.1
    • Smaller values impose tighter priors (more shrinkage)

  • Lambda2: Variable weighting

    • Range: (0, ∞)
    • Default: 0.5
    • Controls relative weights between own-variable and cross-variable lags

  • Lambda3: Lag decay

    • Range: (0, ∞)
    • Default: 1
    • Controls how quickly prior importance decays with lag length

  • Lambda4: Exogenous variable tightness

    • Range: (0, ∞)
    • Default: 100
    • Controls prior tightness on exogenous variable coefficients

  • Lambda5: Block exogeneity shrinkage

    • Range: (0, ∞)
    • Default: 0.001
    • Controls shrinkage when block exogeneity is imposed

Stochastic Volatility Parameters

  • HeteroskedasticityAutoRegression: AR coefficient on residual variance

    • Range: [0, 1)
    • Default: 1
    • Controls persistence in stochastic volatility process

  • HeteroskedasticityShape: IG shape parameter on residual variance

    • Range: (0, ∞)
    • Default: 0.001
    • Controls prior beliefs about volatility distribution shape

  • HeteroskedasticityScale: IG scale parameter on residual variance

    • Range: (0, ∞)
    • Default: 0.001
    • Controls prior beliefs about volatility level

Random Inertia Parameter

  • HeteroskedasticityAutoRegressionVariance: Prior variance of inertia
    • Range: (0, ∞)
    • Default: 0.01
    • Controls the variability of time-varying inertia parameters

Advanced Options

  • Exogenous: Priors on exogenous variables flag

    • Default: false
    • Enables specialized priors for exogenous variables

  • BlockExogenous: Block exogeneity flag

    • Default: false
    • Imposes block exogeneity restrictions

MCMC Settings

  • Burnin: Number of burn-in draws

    • Range: [0, ∞)
    • Default: 0
    • Recommended: Several thousand for proper convergence

  • StabilityThreshold: Threshold for maximum eigenvalue magnitude

    • Range: (0, ∞)
    • Default: null (no stability check)
    • Ensures VAR stability by rejecting draws with eigenvalues above threshold

  • MaxNumUnstableAttempts: Maximum number of unstable sampling attempts

    • Range: [1, ∞)
    • Default: 1000
    • Maximum attempts to find stable draw before stopping

Usage Guidelines

When to Use

  • Time-varying persistence: When the persistence of economic shocks changes over time
  • Policy regime changes: When policy effectiveness varies across periods
  • Structural evolution: When economic relationships exhibit changing dynamics
  • Business cycle analysis: When shock transmission varies across cycle phases

When Not to Use

  • Constant persistence: When shock persistence is stable over time
  • Simple volatility patterns: When basic stochastic volatility is sufficient
  • Short time series: Insufficient data for identifying time-varying inertia
  • Computational constraints: More complex than standard stochastic volatility

Practical Considerations

  • Requires careful specification of inertia variance prior
  • Monitor convergence of both volatility and inertia parameters
  • Consider economic interpretation of time-varying persistence
  • Assess whether added complexity improves model performance

Model Specification

Basic VAR Structure

y_t = X_t β + A^{-1} Σ_t^{1/2} ε_t

where Σ_t follows stochastic volatility.

Stochastic Volatility

log(σ²_{i,t}) = γᵢ log(σ²_{i,t-1}) + η_{i,t}

where γᵢ may vary over time (random inertia).

Random Inertia Component

The inertia parameters follow:

γᵢₜ = γᵢₜ₋₁ + ζᵢₜ

where ζᵢₜ ~ N(0, ζ₀) represents innovations to persistence.

Computational Characteristics

Performance

  • Speed: Moderate to slow due to time-varying inertia estimation
  • Memory: Higher requirements for inertia parameter storage
  • Convergence: Generally stable with proper tuning
  • Complexity: More complex than standard stochastic volatility

Implementation Features

  • Joint estimation of volatility and inertia parameters
  • Specialized MCMC algorithms for time-varying persistence
  • Careful handling of parameter interactions
  • Robust numerical methods for state-space filtering

Comparison with Related Models

vs. Standard Stochastic Volatility

  • Flexibility: Captures time-varying persistence vs. constant persistence
  • Complexity: Additional layer of time-variation
  • Interpretation: Richer understanding of shock transmission dynamics
  • Data requirements: Needs longer series for identification

vs. Time-Varying Parameter VARs

  • Focus: Persistence variation vs. coefficient variation
  • Scope: Volatility and persistence vs. full parameter evolution
  • Efficiency: More efficient when only persistence changes
  • Computational: Less demanding than full time-varying specifications

Economic Interpretation

Inertia Dynamics

  • High inertia periods: Shocks have long-lasting effects
  • Low inertia periods: Shocks dissipate quickly
  • Time-varying inertia: Changing transmission mechanisms

Policy Implications

  • Monetary policy effectiveness may vary with inertia
  • Fiscal multipliers can depend on shock persistence
  • Crisis periods may exhibit different inertia patterns
  • Structural reforms may affect shock transmission

Implementation Considerations

Prior Specification

  • Inertia variance parameter is crucial for identification
  • Balance flexibility with parameter stability
  • Consider economic priors on persistence ranges
  • Monitor sensitivity to hyperparameter choices

Diagnostic Assessment

  • Plot estimated inertia parameters over time
  • Compare with known structural changes
  • Assess economic plausibility of inertia evolution
  • Validate forecasting improvements

Model Validation

Economic Consistency

  • Inertia changes should align with economic theory
  • Compare with regime-dating methodologies
  • Assess relationship with policy changes
  • Evaluate business cycle implications

Statistical Validation

  • Out-of-sample forecasting performance
  • Comparison with nested simpler models
  • Residual diagnostics for remaining autocorrelation
  • Parameter stability assessment

References

  • Carriero, A., T.E. Clark, and M. Marcellino (2019). "Large Vector Autoregressions with Stochastic Volatility and Non-Conjugate Priors"
  • Cogley, T. and T.J. Sargent (2005). "Drifts and Volatilities: Monetary Policies and Outcomes in the Post WWII US"
  • Koop, G. and D. Korobilis (2013). "Large Time-Varying Parameter VARs"
  • Primiceri, G.E. (2005). "Time Varying Structural Vector Autoregressions and Monetary Policy"

BEARX Toolbox Wiki

Clone this wiki locally