Application of Nonlinear Autoregressive Neural Network with Exogen for Modeling Demographic Dynamics in Large Urban Agglomeration

Abstract: This work aims to study the possibility of using artificial neural networks (ANN) for the analysis of time series in special fields. The modeling of demographic dynamics of large urban agglomeration by ANN is conducted on the example of the city of Ekaterinburg, Russia. The prognosis of changes in the population and its sex-age composition, as well as the contribution of the migration factor, are carried out with help of a nonlinear autoregressive exogenous neural network from the Matlab package. The deviation of the forecast results from real data is 3%.