acpbvar - Bayesian Structural VARs with Asymmetric Conjugate Priors

Implements Bayesian estimation of structural Vector Autoregressions in the asymmetric A-form of Chan (2022) <doi:10.3982/QE1381>, with an asymmetric Minnesota prior whose own-lag and cross-lag hyperparameters are selected by empirical Bayes via the closed-form marginal likelihood. Exact equation-by-equation posterior simulation in the structural form (independent Normal-Inverse-Gamma draws) provides posterior draws of the reduced-form parameters; structural identification through sign restrictions follows the algorithm of Rubio-Ramirez, Waggoner and Zha (2010) <doi:10.1111/j.1467-937X.2009.00578.x>. Impulse responses, posterior bands, and acceptance diagnostics are computed from the accepted draws.