Giorgio E. Primiceri

Overview

Giorgio E. Primiceri is a macroeconomist at Northwestern University working on Bayesian time-series econometrics and empirical monetary economics. He is best known for the time-varying-parameter structural VAR with stochastic volatility, and for hierarchical prior selection for Bayesian VARs.

Key Contributions / Features

• TVP-SVAR with stochastic volatility — Primiceri (2005): The foundational model in which both the VAR coefficients and the full innovation covariance matrix drift over time, with a triangular $A_t/\Sigma_t$ decomposition, random-walk laws of motion, and Carter–Kohn + Kim–Shephard–Chib Gibbs estimation; applied to U.S. monetary policy, finding "bad luck" (volatile non-policy shocks) mattered more than policy changes. See Time-Varying Parameter VAR and Primiceri (2005).
• Corrigendum — Del Negro–Primiceri (2015): With Marco Del Negro, corrected the MCMC step-ordering error in the Primiceri (2005) algorithm and gave the general recipe for correctly applying the Kim–Shephard–Chib mixture-of-normals stochastic-volatility procedure. See Del Negro-Primiceri (2015).
• Hierarchical prior selection — Giannone–Lenza–Primiceri (2015): With Domenico Giannone and Michele Lenza, a data-based (hierarchical / empirical-Bayes) approach to setting Bayesian VAR prior informativeness via the marginal likelihood. See Giannone-Lenza-Primiceri (2015) and Minnesota Prior.

Related

• Time-Varying Parameter VAR
• Stochastic Volatility
• Monetary Policy Shocks
• Bayesian VAR
• Primiceri (2005)
• Domenico Giannone