How Good ANN Identification of Post-Stabilization Inflation Dynamics Can Be?

The recent emerging trend in financial systems engineering relies on exploiting soft-computing technologies, and on employing neural-nets techniques, in particular. Simultaneously, recent empirical studies on economic stabilization programs implemented worldwide have clearly demonstrated that, after the successful disinflation, the inflationary process can no longer be captured and explained using the traditional variables and models provided by economic theory. This paper proposes a combined stochastic and artificial neural-nets approach in expert support systems to the identification of inflation dynamics by means of Box-Jenkins ARIMA and Elman-ANN models. The approach is illustrated by means of the case-study data set on inflation dynamics in the pre-and post-stabilization period in the Republic of Macedonia.