金鹽價格預測模型之研究--以U公司PCB事業部為例

Abstract

近年來，由於中國、印度等新興國家經濟建設的高度發展，人民生活水準的提高，促使對能源及原物料的需求急遽增加；然而，原物料的供給量卻未明顯增加，導致需求量高於供給量，依據經濟學理論，勢必造成價格快速上揚，如此一來，將造成企業在取得原物料的成本增加，直接侵蝕掉企業的獲利。 本研究以金鹽(Potassium Gold Cyanide)價格為研究對象，應用時間序列(Time Series)方法，預測國際金價波動方向，透過迴歸分析，瞭解金鹽價格與國際金價具有高度正相關，進而預測金鹽價格未來一個月的走勢。研究期間以2006年7月到2010年12月之國際金價月平均價與金鹽成交月平均價資料作為實證研究資料。本研究分析以移動平均法(Moving Averages，MA)、指數平滑法（Exponential Smoothing，ES）及線性迴歸分析(linear regression)，實證結果顯示以線性迴歸法結果最佳，顯示近年來國際金價漲多跌少，呈顯價格穩定的上揚，利用此採購預測模式，可使採購成本降低約1.28%。 本研究結果顯示，價格預測方法的使用，有助於印刷電路板(Printed Circuit Board，PCB)產業鍍金製程降低金鹽成本，進而推廣至相同製程的產業，用於降低因物料上漲所帶來之資金壓力與購料風險，進而獲得最佳的經營模式。

In recent years, economic high development in the emerging countries such as China, India caused the rapid increase in demand for raw materials and energy resource. However, the supply of raw materials has not increased significantly, resulting in high increase in the price of meterials. This will reduce the profit of firms. In this study we focus on the price of Potassium Gold Cyanide, and use the method of time series to account for the international gold price fluctuations, and find out whether or not high positive correlation exists between the international gold price and the Potassium Gold Cyanide price. We also predict the trend of the Potassium Gold Cyanide price for the upcoming month.This research data is based on the average monthly price of the international gold trading and Potassium Gold Cyanide from July 2006 to December 2010. There are three methods used: the moving average, the exponential smoothing and the linear regression analysis. It is shown that the linear regression is the best method, as it could reduce 1.28% of the cost of procurement, especially when the international gold price rised steadily. The results also show that using price prediction methods can help the printed circuit board industry reduce the cost of Potassium Gold Cyanide. In fact, it can be applied to any of the industries which have the same manufacturing process and need to reduce the risk of purchasing materials due to the high increase in raw materials.