A0637
Title: Predictive model selection for jump diffusion models
Authors: Yuma Uehara - Kansai University (Japan) [presenting]
Abstract: A model selection problem is considered for jump-diffusion models based on high-frequency samples. The terminal time is supposed to diverge (ergodic setting), and the interest is to select drift and diffusion coefficients and jump distribution among candidates. An explicit AIC-type information criterion is proposed based on the threshold quasi-likelihood. Unlike the diffusion case, when the jump term is parametrized in some way, the stochastic flow approach cannot be directly used to get the transition density estimates, which is essential to evaluate the bias. To validate such an approximation, new transition density estimates are presented.
