static_cross_panel

Static Cross-Unit Panel VAR (StaticCrossPanel)

Overview

The Static Cross-Unit Panel VAR explicitly models constant cross-unit correlations and spillovers in panel VAR systems. This estimator captures interdependencies between units through static (time-invariant) cross-unit relationships while maintaining unit-specific dynamics, making it suitable for panel applications where spillover effects between units are important but constant over time.

Methodology

The Static Cross-Unit Panel VAR specification features:

• Unit-specific VAR dynamics with cross-unit dependencies
• Static (time-invariant) spillover effects between units
• Minnesota-type priors on within-unit coefficients
• Explicit modeling of cross-unit error correlations
• Structured approach to inter-unit spillover estimation

This approach extends standard panel VARs by explicitly modeling how units affect each other through constant spillover mechanisms.

Parameters

Cross-Unit Error Parameters

• Alpha0: IG shape parameter on residual variance

  • Range: (0, ∞)
  • Default: 1000
  • Controls prior beliefs about residual variance distribution shape

• Delta0: IG scale parameter on residual variance

  • Range: (0, ∞)
  • Default: 1
  • Controls prior beliefs about residual variance level

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

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 cross-unit model convergence

• StabilityThreshold: Threshold for maximum eigenvalue magnitude

  • Range: (0, ∞)
  • Default: null (no stability check)
  • Ensures stability of full cross-unit system

• MaxNumUnstableAttempts: Maximum number of unstable sampling attempts

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

Usage Guidelines

When to Use

• Cross-unit spillovers: When spillover effects between units are important
• Constant spillovers: When spillover effects are stable over time
• Spatial/network panels: When units are connected through networks or geography
• Policy transmission: When analyzing policy transmission across units

When Not to Use

• Independent units: When units do not significantly affect each other
• Time-varying spillovers: When spillover effects change over time (use Dynamic Cross Panel)
• Large panels: When computational burden becomes prohibitive
• Weak spillover identification: When insufficient data to identify cross-unit effects

Practical Considerations

• Requires sufficient cross-sectional dimension for spillover identification
• Monitor whether spillover effects are economically meaningful
• Consider computational complexity that increases with cross-sectional size
• Assess stability of full system including cross-unit relationships

Model Specification

Panel VAR with Cross-Unit Dependencies

For unit i:

y_{i,t} = Σⱼ A_{ij} y_{j,t-1} + ε_{i,t}

where A_{ij} represents cross-unit spillover coefficients.

Cross-Unit Structure

Spillover matrix A captures static relationships:

A = [A_{11} A_{12} ... A_{1N}]
    [A_{21} A_{22} ... A_{2N}]
    [  ...   ...  ...   ... ]
    [A_{N1} A_{N2} ... A_{NN}]

Error Structure

Cross-unit error correlations:

E[ε_{i,t} ε_{j,t}'] = Ω_{ij}

where Ω captures contemporaneous cross-unit correlations.

Computational Characteristics

Performance

• Speed: Slow due to cross-unit dependencies
• Memory: High memory requirements scaling with N²
• Convergence: More challenging than separable panel models
• Scalability: Limited scalability with number of units

Cross-Unit Complexity

• Joint estimation of all cross-unit relationships
• Large parameter space for spillover effects
• Computational burden increases rapidly with panel size
• Requires careful MCMC tuning

Cross-Unit Spillover Analysis

Spillover Identification

• Within-unit effects: A_{ii} coefficients
• Cross-unit spillovers: A_{ij} coefficients (i ≠ j)
• Network effects: Pattern of significant A_{ij}
• Spillover persistence: Dynamic properties of full system

Economic Interpretation

• Direct effects: Impact on own-unit variables
• Spillover effects: Impact on other units
• Network centrality: Units with strong outward spillovers
• Vulnerability: Units highly affected by others

Comparison with Alternative Panel Approaches

vs. Separable Panel Models

• Cross-unit effects: Explicitly models vs. ignores spillovers
• Complexity: Much more complex estimation
• Information: Captures cross-unit relationships vs. pooling only
• Applications: Suited for interconnected vs. independent units

vs. Dynamic Cross Panel

• Time variation: Static vs. time-varying spillovers
• Complexity: Less complex than dynamic spillovers
• Assumptions: Constant spillovers vs. evolving relationships
• Data requirements: Less demanding than dynamic specification

Implementation Strategy

Model Building

• Start with separable panel models to understand within-unit dynamics
• Gradually introduce cross-unit relationships
• Consider economic theory to guide spillover structure
• Use shrinkage priors on spillover coefficients

Computational Management

• Monitor convergence across all cross-unit parameters
• Use block sampling for different parameter groups
• Consider parallelization opportunities
• Plan for substantial computational resources

Model Assessment

Spillover Validation

• Test economic significance of estimated spillovers
• Compare with known network/geographic relationships
• Assess spillover magnitude vs. within-unit effects
• Validate stability of full cross-unit system

Performance Evaluation

• Compare forecasting performance with separable models
• Assess improvement from incorporating spillovers
• Evaluate computational cost vs. performance gains
• Cross-validate spillover estimates across subsamples

Economic Applications

Suitable Contexts

• International trade and spillovers
• Regional economic interdependencies
• Financial contagion analysis
• Policy transmission across jurisdictions

Specific Use Cases

• Monetary policy spillovers across countries
• Fiscal multipliers in currency unions
• Financial market interconnections
• Supply chain disruption propagation

References

• Canova, F. and M. Ciccarelli (2009). "Estimating Multi-Country VAR Models"
• Pesaran, M.H., T. Schuermann, and S.M. Weiner (2004). "Modeling Regional Interdependencies using a Global Error-Correcting Macroeconometric Model"
• Diebold, F.X. and K. Yilmaz (2014). "On the Network Topology of Variance Decompositions: Measuring the Connectedness of Financial Firms"
• Barigozzi, M. and M. Brownlees (2019). "NETS: Network Estimation for Time Series"