Bayesian Inference in Stochastic Form of Multiplicative Error Model

In this paper, a stochastic form of multiplicative error model is presented by considering the long term trend as the process of a state space model. Optimal filtering technique via sequential Monte Carlo perspective is developed for tracking the long term trend as the hidden state of this model and the Bayesian framework is applied to the estimation of both the hidden state and the unknown parameters using particle marginal Metropolis Hastings scheme. A real world application is also performed to illustrate and evaluate the proposed methodology.