A0246
Title: Bias-reduction in state-space model estimation
Authors: Magda Monteiro - University of Aveiro (Portugal) [presenting]
Marco Costa - University of Aveiro (Portugal)
Abstract: State-space models (SSM) provide a versatile framework for representing dynamic systems, where latent states evolve through a stochastic process and observed data are connected via a linear measurement equation. A method is proposed for estimating the parameters of SSM, based on a double-iterated generalized method of moments (GMM) procedure. The proposed approach is designed to reduce estimation bias, particularly in contexts involving small sample sizes or high levels of variability. Simulation results demonstrate that this method outperforms traditional estimators, such as maximum likelihood, in terms of accuracy and robustness.
