The prediction of Iran's defensive expenditures up to 2025 horizon using genetic algorithms and PSO

The defense sector is one of the key sectors of Iran that has devoted an important part of the total public spending. Any effort for forecasting the future defense expenditure trend is of special importance for policy-makers of defense sector. In this paper, we specified a defense expenditure model relying on theoretical basics in order to obtain desirable forecasts. On the basis of three forms of linear, exponential and quadratic equations and using theoretical foundations in the field of defense expenditure function, we used genetic algorithm (GA) and particle swarm optimization (PSO) algorithm to simulate Iranians defense expenditure during 1959-2012. Then we selected the superior model in terms of prediction power criteria and forecast consumption until 2025. Based on the results obtained, it can be said that the defense cost function simulated by the PSO algorithm with exponential form is more compatible with the structure of Iran and therefore, it has been selected to predict expenditures from the sample until 2025. The predicted results also indicate that defense spending in Iran will increase with a relatively moderate slope by 2025.