Computational Modeling of ARIMA-based G-MFS Methods: Long-term Forecasting of Increasing Population

Every year the population in each region has increased, no exception in the province of NTB. Therefore, to take the policies of upgrading the quality of life of residents in NTB province it is necessary to predict the number of people in the future as a benchmark decision making by the Government. The method used is ARIMA. This research aims to construct the G-MFS-based mathematical model on the ARIMA method which is then continued to determine the number of population predictions in NTB province for the next few years. The data used is the population data in 10 districts in the province of NTB for the last 11 years. Based on the results of the simulation found that of the four mathematical models that became the output of the G-MFS turned out to be the 4th model gives the highest level of accuracy on each data simulated with an average MSE of 0207, while the increase in population in the year 2020 average of 1.15% and for the next 20 years the average increase of 0.62%. The forecasting of the population is expected to provide an important contribution to the Government as a weighting material in the planning, implementation, repair, and formulation of subsequent policies, both in the field of economy, education, health, and reducing the level of poverty in the NTB province in the future.

Key words : ARIMA method; GUI Multimodel Forecasting System; Population forecasting