Modeling and Forecasting GDP Growth rate based on Population Growth Scenarios; Using Neural Networks

Beyond the pessimistic perspective of the impact of population growth on economic growth, institutional approach which is in contrast with the mainstream neoclassical growth models, have presented different and somehow positive results. In this paper, we used hybrid GMDH neural networks and genetic algorithms, based on two variables: population growth and fertility, to model and predict GDP growth. The results showed that first; population growth has considerable effects on economic growth. Second, the economic growth is the short term (and non-linear) result of population growth. Moreover, GDP growth has been forecasted based on various scenarios of population growth and total fertility. The finding indicated that the best forecasting is related to low population scenario and slow increase total fertility scenario.