Title: Model comparison for ergodic SDEs in YUIMA
Authors: Shoichi Eguchi - Osaka Institute of Technology (Japan) [presenting]
Abstract: There are several studies of model selection for stochastic differential equations (SDEs), which include the contrast-based information criterion for ergodic diffusion processes, AIC-type information criterion for ergodic Levy-driven SDEs, and the Schwarz-type information criterion for locally asymptotically quadratic models. Based on these studies, in R package yuima, the function for model selection for diffusion processes has been implemented. However, this function is not compatible with the model selection for Levy-driven SDEs. We will overview the model selection methods for Levy-driven SDEs and explain the improvements of the model selection function.