gen_large_shock_sv

General Stochastic Volatility VAR for Large Shocks (GenLargeShockSV)

Overview

The General Large Shock Stochastic Volatility VAR represents an extended version of the Large Shock SV framework with additional flexibility for handling extreme events. This estimator provides the most general specification for modeling large shocks and volatility changes, combining sophisticated scaling factor mechanisms with enhanced parameter flexibility for complex crisis modeling scenarios.

Methodology

The GenLargeShockSV specification features:

• Constant VAR coefficients with Minnesota-type priors
• Advanced stochastic volatility with scaling factors for large shocks
• Extended Pareto-distributed scaling factors with enhanced flexibility
• Generalized time-varying persistence in volatility scaling
• Most flexible large shock modeling framework available

This approach builds upon the Large Shock SV framework while providing additional generality for complex volatility patterns and extreme event modeling.

Parameters

Large Shock Parameters

• Mult0: Initial mean of scaling factors

  • Type: Vector of numbers
  • Default: "" (empty, estimated)
  • Controls initial beliefs about volatility scaling magnitude

• ScaleMult: Scale parameter for Pareto distribution of scaling factors

  • Type: Vector of numbers
  • Default: "" (empty, estimated)
  • Controls the scale of the Pareto distribution for volatility multipliers

• ShapeMult: Shape parameter for Pareto distribution of scaling factors

  • Type: Vector of numbers
  • Default: "" (empty, estimated)
  • Controls the tail behavior of large shock distribution

• PropStdMult: Proposal standard deviation for scaling factors

  • Type: Vector of numbers
  • Default: "" (empty, calibrated)
  • MCMC tuning parameter for scaling factor updates

Enhanced Autoregressive Scaling Parameters

• MultAR0: Initial mean for scaling factor AR parameter

  • Range: [0, 1]
  • Default: 0.5
  • Controls initial persistence in volatility scaling

• AlphaMultAR: Alpha parameter for beta distribution of AR coefficient

  • Type: Number
  • Default: "" (empty, estimated)
  • Beta distribution shape parameter for AR coefficient

• BetaMultAR: Beta parameter for beta distribution of AR coefficient

  • Type: Number
  • Default: "" (empty, estimated)
  • Beta distribution shape parameter for AR coefficient

• PropStdAR: Proposal standard deviation for AR parameter

  • Type: Number
  • Default: "" (empty, calibrated)
  • MCMC tuning parameter for AR coefficient updates

Crisis Period Settings

• Turningpoint: Start date of high-volatility period
  • Type: Date
  • Default: "" (empty)
  • Specifies when the large shock period begins

Minnesota-Style Prior Parameters (Active)

• 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)

• Lambda3: Lag decay

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

MCMC Settings

• Burnin: Number of burn-in draws

  • Range: [0, ∞)
  • Default: 0
  • Recommended: Extensive burn-in due to complex volatility structure

• 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

• Complex crisis modeling: When standard large shock models are insufficient
• Multiple crisis types: When different types of extreme events occur
• Enhanced flexibility: When maximum flexibility in shock modeling is needed
• Research applications: When exploring complex volatility dynamics

When Not to Use

• Simple applications: When basic stochastic volatility is adequate
• Computational limitations: Among the most demanding estimators computationally
• Small samples: Requires very long time series for identification
• Parsimony concerns: When model simplicity is prioritized

Practical Considerations

• Requires extremely long time series for reliable estimation
• Computationally intensive beyond standard large shock models
• Extensive parameter tuning and convergence monitoring essential
• Consider whether enhanced complexity provides meaningful improvements

Model Specification

Enhanced Structure

y_t = X_t β + A^{-1} Σ_t^{1/2} M_t^{1/2} ε_t

where M_t represents enhanced time-varying scaling factors for large shocks.

Generalized Scaling Factor Process

M_{i,t} ~ Pareto(scale_i, shape_i) with enhanced flexibility
log(M_{i,t}) = ρ_i log(M_{i,t-1}) + ν_{i,t}

where ρ_i follows a generalized Beta distribution with additional parameters.

Advanced Volatility Evolution

Standard stochastic volatility combined with generalized large shock scaling:

log(σ²_{i,t}) = log(σ²_{i,t-1}) + η_{i,t}
Var(ε_{i,t}) = σ²_{i,t} × M_{i,t}

with enhanced modeling of M_{i,t} dynamics.

Computational Characteristics

Performance

• Speed: Extremely slow due to enhanced complexity
• Memory: Very high memory requirements for extended parameter space
• Convergence: Challenging convergence requiring specialized algorithms
• Numerical stability: Requires advanced numerical methods

Enhanced Implementation Features

• Extended parameter space requires sophisticated sampling
• Advanced scaling factor dynamics
• Generalized crisis period modeling
• Maximum flexibility in shock accommodation

Comparison with Related Models

vs. Large Shock SV

• Generalization: More general specification with enhanced flexibility
• Parameters: Extended parameter space for complex modeling
• Performance: Potentially better for very complex volatility patterns
• Complexity: Significantly more complex estimation

vs. Standard Stochastic Volatility

• Robustness: Maximum robustness to extreme observations
• Complexity: Extreme complexity vs. basic specifications
• Applications: Specialized for complex crisis modeling
• Requirements: Very demanding data and computational requirements

Implementation Strategy

Advanced Initialization

• Use simpler large shock models for starting values
• Sequential parameter introduction approach
• Multiple chain strategy with extensive overdispersion
• Adaptive MCMC methods for complex parameter space

Specialized Tuning

• Enhanced Pareto parameter calibration
• Generalized AR parameter tuning
• Crisis period specification requires domain expertise
• Extensive sensitivity analysis across parameter dimensions

Model Validation

Comprehensive Testing

• Compare with nested simpler large shock specifications
• Extensive out-of-sample evaluation across crisis periods
• Cross-validation against known extreme events
• Economic interpretation of enhanced parameter estimates

Advanced Diagnostics

• Multi-dimensional convergence assessment
• Parameter interaction analysis
• Enhanced scaling factor interpretation
• Crisis dating validation

Economic Applications

Specialized Use Cases

• Financial crisis analysis with multiple shock types
• Sovereign debt crisis modeling
• Currency crisis with complex volatility patterns
• Academic research on extreme event modeling

Policy Analysis

• Central bank stress testing scenarios
• Macroprudential policy evaluation during crises
• Systemic risk assessment with complex shocks
• International spillover analysis during extreme events

References

• Carriero, A., T.E. Clark, and M. Marcellino (2019). "Large Vector Autoregressions with Stochastic Volatility and Non-Conjugate Priors"
• Carriero, A., G. Kapetanios, and M. Marcellino (2012). "Forecasting Large Datasets with Bayesian Reduced Rank Multivariate Models"
• Adrian, T. and H.S. Shin (2014). "Procyclical Leverage and Value-at-Risk"
• Brunnermeier, M.K. and Y. Sannikov (2014). "A Macroeconomic Model with a Financial Sector"