189 – Contributed Oral Poster Presentations: General Methodology
Quasi-Likelihood Estimation on SFIEGARCH Process: A Simulation Study
Silvia Lopes
Federal University of Rio Grande do Sul
Taiane Prass
Federal University of Rio Grande do Sul
The theoretical properties of the quasi-maximum likelihood estimate (QMLE) for SFIEGARCH models are still an open issue. This work presents a simulation study to access the finite sample performance of this estimator for misspecified SFIEGARCH(0; d; 0)s process. The simulation study presented here considers the Student's t and the generalized error distribution (GED) as underlying distributions for the innovation process. In order to obtain lighter and heavier tails than the Gaussian distribution, different values for the nuisance parameter are considered for both underlying distributions.