Smooth Threshold Autoregressive models and Markov process: An application to the Lebanese GDP growth rate

Abstract

This paper analyzes the evolution of the Lebanese GDP growth rate over the period 1970-2019 by estimating two kinds of switching models: The Smooth Transition Autoregressive (STAR) model and the model of the Markov process. These models show, on the one hand, asymmetries in the evolution of GDP growth with an abrupt transition from a regime to another and, on the other hand, a high probability that the economy remains in the recession regime. Even though the duration of the expansion phase is longer than the duration of the recession phase, the Lebanese economy experiencing the greatest difficulties in moving from a recession regime to an expansion regime. In addition, such an evolution is explosive and volatile during the lower regime (recession phase) but stationary and damped in the upper regime (expansion phase). Finally, the paper shows that the STAR model, taking a logistic form, better fits the Lebanese GDP growth than the Markov model.