Application of Artificial Neural Network to the Predictability of Social Welfare and Economic Growth on Defense Expenditure—The Case of Taiwan and China

• Main Authors: Ching-Hua Chen, 陳慶華
• Other Authors: Mei-Hui Chen
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/3876k2

碩士 === 國防大學管理學院 === 財務管理學系 === 101 === The situation between Taiwan and China has improved markedly in recent years. In view of the increase in China's defense spending, Taiwan’s defense spending, as a percentage of its GDP, has been steadily decreasing over the past several years, indicating that Taiwan’s defense policy has changed from an offensive position to a defensive-oriented stance. However, the military China-Taiwan tensions are still not been reduced. Therefore, the prediction of Taiwan’s defense spending can provide insights on Taiwan’s defense policy and military strategy. In this study, I apply ANNs modeling, and MAPE as well as MAE are used to evaluate the forecast accuracy of Taiwan’s defense expenditure. The lagged variables of social welfare budget and the economic growth are used as the inputs of ANNs modeling. The results show that the forecasting performance of ANN is in high accuracy.