A1787
Title: Efficient parameter estimation for multivariate jump-diffusions
Authors: Gustavo Schwenkler - Boston University (United States) [presenting]
Francois Guay - Cornerstone Research (United States)
Abstract: Unbiased estimators of the transition density and posterior filters of a multivariate jump-diffusion process are developed. The drift, volatility, jump intensity, and jump magnitude are allowed to be state-dependent and non-affine. It is not necessary to diagonalize the volatility matrix. Our approach facilitates the parametric estimation of multivariate jump-diffusion models based on discretely observed data. Our parameter estimators have the same asymptotic behavior as maximum likelihood estimators under mild conditions. Our methodology is found to be highly accurate and computationally efficient for the estimation of consumption growth dynamics.
