Shawn Ni

Overview

Shawn Ni is an economist at the University of Missouri, Columbia. His research spans Bayesian econometrics, VAR models, and consumption economics.

Sun-Ni (2005): Co-developed the comparison of constant-Jeffreys vs. constant-reference priors for Bayesian VAR estimation, establishing the over-estimation bias of the Jeffreys prior and the practical superiority of the Yang-Berger reference prior. See Noninformative Prior for VAR.
Applied a 10-variable consumption-growth VAR to quarterly US data (1967Q1–2000Q3) demonstrating the practical magnitude of the prior difference.
Ni-Sun (2005): Extended the Bayesian VAR comparison to include a shrinkage prior on the coefficient matrix $\Phi$ ( $\pi_S(\phi) \propto \|\phi\|^{-(J-2)}$ , implemented via a two-level hierarchy), LINEX asymmetric loss ( $a_{ij} = -4$ for non-intercept coefficients), and a Student-t error model. Established that prior choice for Φ dominates loss function choice in frequentist risk; shrinkage+reference combination best overall. U.S. macro application: GDP-to-inflation IRF changes sign under shrinkage prior. See Noninformative Prior for VAR.
George-Sun-Ni (2008): Co-developed Bayesian stochastic search variable selection for VAR models, extending SSVS to simultaneously identify restrictions on the coefficient matrix Φ and the Cholesky precision factor Ψ. Five-step all-standard-form Gibbs sampler; Rao-Blackwell forecasts improve over MLE by 20–67% in simulation; empirical application to 7-variable PPI→CPI VAR identifying supply-chain contemporaneous structure and post-1981 structural change. Journal of Econometrics 142 (2008): 553–580.