ordinary

Ordinary Least Squares (Ordinary)

Description

Classical ordinary least squares (OLS) estimation without priors. This is the baseline VAR estimator that provides maximum likelihood estimates through equation-by-equation OLS regression.

Technical Details

• Uses equation-by-equation OLS estimation
• No prior information incorporated
• Posterior sampling available for uncertainty quantification
• Can be used as benchmark for comparison with Bayesian estimators

User Settings (Hyperparameters)

Core Settings

FixedBeta (logical)

• Description: Keep VAR coefficients fixed when sampling from posterior distribution
• Default: false
• Details: When set to true, the coefficient matrix β is held constant at its OLS estimate during posterior sampling, eliminating parameter uncertainty for coefficients

FixedSigma (logical)

• Description: Keep error covariance matrix fixed when sampling from posterior distribution
• Default: false
• Details: When set to true, the error covariance matrix Σ is held constant at its OLS estimate during posterior sampling, eliminating parameter uncertainty for error variances

Inherited Base Settings

Burnin (double)

• Description: Number of burn-in draws for posterior sampling
• Default: 0
• Details: Number of initial MCMC draws to discard before collecting posterior samples

StabilityThreshold (double)

• Description: Threshold for maximum eigenvalue magnitude to ensure VAR stability
• Default: Inf (no stability check)
• Details: Maximum allowed eigenvalue magnitude; draws with eigenvalues exceeding this are rejected

MaxNumUnstableAttempts (double)

• Description: Maximum number of unstable sampling attempts before giving up
• Default: 1000
• Details: If stability checking is enabled, this limits the number of consecutive unstable draws before termination

Usage Notes

• Primary use case is as a benchmark for comparison with Bayesian estimators
• Provides classical statistical inference through asymptotic theory
• No prior information incorporated, so results depend entirely on sample data
• Only has two specific settings (FixedBeta and FixedSigma) plus standard sampling control parameters
• When both FixedBeta and FixedSigma are true, posterior sampling reduces to point estimates