A0512
Title: Asymptotic inference for skewed stable Ornstein-Uhlenbeck process
Authors: Hiroki Masuda - University of Tokyo (Japan) [presenting]
Eitaro Kawamo - Kyushu University (Japan)
Abstract: Asymptotic inference is considered for a non-Gaussian stable Levy process with possibly asymmetric jumps. Based on the infill sampling scheme in a fixed period, the local likelihood asymptotics through a non-diagonal normalizing matrix are presented. Moreover, to estimate the parameters in the driving skewed stable noise, an easy-to-compute empirical moment fitting is proposed. This can serve as an initial estimator for the numerical search of the maximum-likelihood estimator, allowing the bypass of the heavy computational load in optimizing the original log-likelihood function. Simulation results are given to show that the proposed estimator achieves accuracy comparable to that of the maximum likelihood estimator in a much shorter time.
