Overview
Luc Bauwens is an econometrician at CORE (Center for Operations Research and Econometrics), Université catholique de Louvain, Belgium. He is known for Bayesian inference in simultaneous equations and cointegrated VAR systems, GARCH modelling, and for the textbook treatment of Bayesian inference in dynamic econometric models (Bauwens, Lubrano, and Richard 1999).
Key Contributions / Features
- Bauwens and Lubrano (1994a) — "Identification Restrictions and Posterior Densities in Cointegrated Gaussian VAR Systems": derived the marginal posterior of the cointegrating matrix β under linear identification restrictions; basis for subsequent Bayesian VECM estimation.
- Bauwens and Giot (1997) — "A Gibbs Sampling Approach to Cointegration": applied Griddy-Gibbs sampling to the cointegrated VAR; spectral convergence diagnostics; blocked sampler for correlated parameters.
- Bauwens, Lubrano, and Richard (1999) — Bayesian Inference in Dynamic Econometric Models (Oxford University Press): standard reference text on Bayesian methods for VAR, GARCH, and simultaneous equations models.
- Bauwens, Deprins, and Vandeuren (1998) — "Modelling Interest Rates with a Cointegrated VAR-GARCH Model": joint VECM-BEKK estimation for long and short rates of five OECD countries; demonstrates that GARCH heteroscedasticity sharpens cointegration test power; finds multivariate IGARCH and common volatility factor across rates.
- Bauwens and Rombouts (2007) — "Bayesian Inference for the Mixed Conditional Heteroskedasticity Model", The Econometrics Journal 10(2): 408–425: Gibbs sampler for the MN-GARCH model (Haas-Mittnik-Paolella 2004a); griddy-Gibbs for GARCH parameters, Dirichlet conjugate for mixing weights, Laplace marginal likelihood for K selection; S&P 500 application showing near-IGARCH as a misspecification artifact from ignoring the two-component mixture structure.
- Bauwens, Laurent, and Rombouts (2006) — "Multivariate GARCH Models: A Survey", Journal of Applied Econometrics 21(1): 79–109. Canonical survey of the MGARCH field organized via a three-family taxonomy: (i) direct generalizations (VEC, BEKK, Factor GARCH); (ii) linear combinations (O-GARCH, GO-GARCH); (iii) nonlinear combinations (CCC, DCC, GDC, Copula-GARCH). Covers parameter counts, invariance, QML theory (Jeantheau 1998 consistency; Comte-Lieberman 2003 normality for BEKK), two-step DCC estimation, variance targeting, diagnostic tests (Hosking 1980; Ling-Li 1997; Tse 2000, 2002; Engle-Sheppard 2001), and ten open research questions.
Related