A neural network measurement of relative military security - The case of Greece and Cyprus

Abstract

This paper aims at introducing a relative security measure, applicable to evaluating the impact of arms races on the military security of allies. This measure is based on demographic criteria, which play a dominant role in a number of arms races involving military alliances. The case of Greece and Cyprus, on one hand, and Turkey on the other, is the one to which our relative security measure is applied and tested. Artificial neural networks were trained to forecast the future behaviour of relative security. The high forecasting performance permitted the application of alternative scenarios for predicting the impact of the Greek-Turkish arms race on the relative security of the Greek-Cypriot alliance.