Eric Jacquier is a finance professor at Boston College's Carroll School of Management and a CIRANO research fellow. His research focuses on Bayesian and simulation-based methods for financial econometrics, including stochastic volatility estimation and contingent claim model error.
Key Contributions / Features
Jacquier, Polson, and Rossi (1994): Introduced Bayesian MCMC estimation for discrete-time stochastic volatility models; single-move Metropolis-within-Gibbs; showed Bayesian MCMC outperforms quasi-maximum likelihood for SV parameters.
Jacquier and Jarrow (2000): Formally incorporated parameter uncertainty and model error into option pricing estimation; Metropolis-Gibbs MCMC for Black-Scholes; fit vs. predictive density distinction; non-parametric polynomial extensions (Jarrow-Rudd); heteroskedastic error structures; showed estimation method is as important as model choice.
Jacquier, Kane, and Marcus (2004): Derived minimum-MSE estimator of long-run expected portfolio value; showed arithmetic mean compounding is catastrophically imprecise for long horizons; proved optimal risky-asset allocation decreases with investment horizon when parameter uncertainty is properly incorporated.
Jacquier, Polson, and Rossi (2004): Extended the JPR (1994) Bayesian MCMC framework to fat-tailed SV (Student-t via scale mixing) and correlated errors (ASV2: corr(ut,vt)=ρ). Bayes Factors decisively favor both extensions for U.S. equity; leverage (ρ<0) dominates fat tails for equity; fat tails smooth volatility path by absorbing outliers. See Stochastic Volatility.
Jacquier, Johannes, and Polson (2007): MCMC-ML algorithm for latent state models — J-copy data augmentation concentrates the parameter marginal on the MLE; J(draws−MLE)→N(0,I−1); applied to SV and multivariate Merton jump-diffusion. See MCMC Maximum Likelihood.
Jacquier and Polson (2010): 92-page handbook survey of Bayesian methods in finance covering portfolio optimization under estimation risk (w* horizon formula), return predictability (Bayesian model averaging), APT factor pricing (McCulloch-Rossi, Geweke-Zhou), SV/GARCH/options (Savage density ratio Bayes factor for leverage), and particle filtering with parameter learning (CJLP 2010). See Jacquier and Polson (2010).