random_inertia_sv_favar

Random Inertia Stochastic Volatility FAVAR (RandomInertiaSVFAVAR)

Overview

The Random Inertia Stochastic Volatility FAVAR combines factor-augmented VAR analysis with Random Inertia Stochastic Volatility modeling in a two-step framework. This estimator first extracts factors via principal components, then applies random inertia stochastic volatility priors to model time-varying volatility in the factor-augmented VAR, allowing for periods of high and low volatility persistence.

Methodology

The Random Inertia SV FAVAR specification features:

• Two-step estimation: factor extraction followed by SV-VAR estimation
• Principal component factor extraction in first step
• Random inertia stochastic volatility modeling with time-varying persistence
• Regime-switching volatility persistence parameters
• Minnesota-style priors combined with stochastic volatility structure

This approach captures both common factor dynamics and regime-dependent volatility patterns with varying persistence.

Parameters

Prior Configuration

• Sigma: Method of calculating priors on covariance matrix
  • Values: "ar" (autoregressive) or "eye" (identity-based)
  • Default: "ar"
  • Controls how the prior covariance matrix is constructed

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

Random Inertia SV Parameters

• SVVolPrior: Prior variance for initial volatility

  • Range: (0, ∞)
  • Default: 10000
  • Controls prior belief about initial volatility levels

• SVVolPriorDims: Prior for volatility of volatility parameters

  • Range: (0, ∞)
  • Default: 2
  • Controls tightness of priors on volatility of volatility

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: 5000
  • Required burn-in for stochastic volatility sampling

• It: Total number of iterations

  • Range: [1, ∞)
  • Default: 20000
  • Total MCMC iterations including burn-in

• 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

• Regime-dependent volatility: When volatility persistence varies over time
• Financial applications: When modeling assets with changing volatility regimes
• Crisis modeling: When volatility clustering varies across different periods
• Factor volatility analysis: When common factors exhibit regime-switching volatility

When Not to Use

• Constant volatility assumed: When volatility is believed to be stable
• Small sample sizes: When insufficient data for regime identification
• Simple applications: When constant volatility SV models are sufficient
• High-frequency data: When simpler volatility models are more appropriate

Practical Considerations

• Requires sufficient time series length for regime identification
• More complex than standard stochastic volatility approaches
• Benefits from periods with clearly different volatility characteristics
• May require careful initialization of regime parameters

Model Specification

Two-Step Procedure

Step 1: Factor Extraction

F_t = Λ X_t + e_t

Factors extracted via principal components from observable dataset X_t.

Step 2: Random Inertia SV FAVAR

[Y_t]
[F_t] = A [Y_{t-1}] + Σ_t^{1/2} ε_t
      [F_{t-1}]

Random Inertia Stochastic Volatility Structure

log σ²_{i,t} = log σ²_{i,t-1} + φ_{i,t} * (log h²_i - log σ²_{i,t-1}) + ω_{i,t}
φ_{i,t} ~ Uniform(φ_{i,low}, φ_{i,high})

where φ_{i,t} represents time-varying persistence in volatility.

Regime-Switching Persistence

• Volatility persistence φ_{i,t} switches between different values
• Transition probabilities govern regime changes
• Allows for periods of high and low volatility persistence
• Captures changing volatility clustering patterns

Computational Characteristics

Performance

• Speed: Slower than constant volatility models due to regime-switching complexity
• Memory: Higher requirements for storing regime states and volatility paths
• Convergence: Moderate convergence, may require longer chains for regime identification
• Scalability: Scales moderately with system size and sample length

MCMC Structure

• Requires sampling of regime indicators and persistence parameters
• Volatility paths sampled conditional on regime states
• VAR coefficients sampled conditional on volatility structure
• May benefit from specialized sampling schemes for regime parameters

Factor-Augmented Analysis

Factor Volatility Regimes

• Common factors can exhibit different volatility regimes
• Regime changes may be synchronized across factors or idiosyncratic
• Factor volatility regimes capture aggregate uncertainty periods
• Useful for identifying crisis periods and uncertainty episodes

Economic Interpretation

• High persistence regimes correspond to prolonged volatility clustering
• Low persistence regimes allow for quick volatility mean reversion
• Regime changes capture structural changes in uncertainty dynamics
• Factor regimes may correlate with business cycle phases

Comparison with Related Approaches

vs. Standard SV FAVAR

• Persistence: Time-varying vs. constant volatility persistence
• Regimes: Captures regime-switching vs. single volatility process
• Flexibility: More flexible but more complex structure
• Applications: Better for data with clear regime changes

vs. CCMM SV FAVAR

• Persistence modeling: Random inertia vs. continuous persistence evolution
• Regime structure: Discrete vs. smooth persistence changes
• Identification: May be easier to identify discrete regimes
• Interpretation: Clearer regime interpretation

vs. Time-Varying FAVAR

• Focus: Volatility regimes vs. coefficient time-variation
• Structure: Stochastic volatility vs. time-varying parameters
• Applications: Volatility analysis vs. parameter instability
• Complexity: Similar computational complexity

Implementation Strategy

Factor Extraction Stage

• Standardize data and handle missing observations
• Select number of factors using information criteria
• Validate factor stability across potential regime periods
• Assess factor interpretability and economic meaning

Random Inertia SV Estimation

• Initialize regime states and persistence parameters carefully
• Use efficient sampling schemes for regime-switching parameters
• Monitor mixing and convergence of regime indicators
• Validate identification of different persistence regimes

Regime Identification

• Use multiple starting points for regime parameters
• Assess regime classification stability
• Compare regime dates with known economic events
• Validate regime persistence and transition patterns

Model Validation

Regime Assessment

• Examine identified regime periods and their economic interpretation
• Test regime stability and transition frequency
• Compare regime classifications across different model specifications
• Assess regime predictability and economic significance

Volatility Analysis

• Examine volatility paths and their regime dependence
• Compare in-sample volatility fit with alternative specifications
• Assess out-of-sample volatility forecasting performance
• Test volatility regime persistence and mean reversion

Factor Performance

• Evaluate factor extraction robustness across regime periods
• Assess factor loadings stability during different volatility regimes
• Compare factor-based forecasting across regimes
• Test structural interpretation consistency

Applications

Financial Risk Management

• Portfolio risk analysis with regime-dependent factor volatility
• Stress testing under different volatility persistence regimes
• Options pricing with factor-based volatility regimes
• Systemic risk assessment through factor volatility clustering

Macroeconomic Analysis

• Business cycle analysis with time-varying uncertainty
• Central bank policy analysis under different volatility regimes
• International spillover analysis with regime-dependent transmission
• Crisis prediction using factor volatility regime indicators

Crisis and Uncertainty Analysis

• Financial crisis identification through factor volatility regimes
• Economic policy uncertainty measurement
• Market sentiment analysis using volatility persistence changes
• Early warning systems based on regime-switching patterns

Advantages and Limitations

Key Advantages

• Captures regime-dependent volatility persistence patterns
• Provides clear economic interpretation of volatility regimes
• Flexible framework accommodating different persistence patterns
• Useful for crisis identification and uncertainty analysis

Main Limitations

• Requires sufficient data for reliable regime identification
• More complex than standard stochastic volatility approaches
• May suffer from overfitting with too many regimes
• Regime identification can be sensitive to model specification

References

• Bernanke, B.S., J. Boivin, and P. Eliasz (2005). "Measuring the Effects of Monetary Policy: A Factor-Augmented Vector Autoregressive (FAVAR) Approach"
• Eisenstat, E., J.C.C. Chan, and R.W. Strachan (2016). "Stochastic Model Specification Search for Time-Varying Parameter VARs"
• Primiceri, G.E. (2005). "Time Varying Structural Vector Autoregressions and Monetary Policy"
• Jacquier, E., N.G. Polson, and P.E. Rossi (2004). "Bayesian Analysis of Stochastic Volatility Models with Fat-Tails and Correlated Errors"