The Forecast Study of Defense Spending of China

• Main Authors: Wen-Cheng, Lin, 林文政
• Other Authors: Kuo-Cheng, Kuo, Wei-Kang, Wang
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/72566564518206811716

碩士 === 國防管理學院 === 國防財務資源研究所 === 96 === China has advanced by leaps and bounds economically in recent ten years; the multiple of the defence spending grows up, cause Asia, global economy and national defence to produce the crisis safely. The international monetary fund will estimate China with first of the posture in the world of GDP rate of increase in advance; become world economic development most heavy to push away hand. So, what a way of prediction developing effectively, with more accurate prediction Chinese defence spending, as the reference that the national defence of our country is administrative, permit it for the present important subject. This research is applied to the prediction of the Chinese defence spending with one kind of neural networks, and pass the Mean Absolute Percentage Error (MAPE) Assess its efficiency in indicator. The result of study shows the model of prediction taking China's previous defence spending and GDP gross output value as parameter, predict the result is the best. In addition, such as as for accuracy of prediction within a short time, middle period, medium and long-term five years, it is the lowest with the average error rate of the way three, show that if predicts within lasting five years with many variables of one kind of neural networks, BPN taking China's previous defence spending, GDP gross output value and inflation rate as parameter predicts models, predict the result is the best.