A Comparison of the Markov Chains, Artificial Neural Networks, and ARIMA for Forecasting of Iraq's Population

Abstract

Statistical methods play an important role in building models that predict the size of the population and thus they help in setting economic and social development plans.This study aims to compare between Markov chains, artificial neural networks, and ARIMA to predict the population of Iraq, based on the data of the population of Iraq for the period 1977-2007.By comparing these models using the criteria of MAE and RMSE, it was concluded that the ARIMA model (1,1,1) is the best model to get on accurate predictions. based on this model, The population of Iraq were predicted until 2030.