應用灰色理論預測台灣糖果產業銷售量之研究

本研究的主要目的，利用灰色理論建立銷售量預測模式，了解目前台灣糖果產業銷售量情形，作為建立銷售量預測模式之依據。研究範圍為2000年至2011年台灣糖果的銷售量，應用灰色理論中GM（1,1）模型、四種轉化式及傳統預測方法中指數平滑法、一般線性分析來做預測分析。最後以平均絕對百分比誤差（MAPE）來衡量模型的績效，以判斷預測值的準確程度，找出最佳的預測方法。研究結果顯示GM（1,1）模型無論在MAPE或是精確度上，皆優於指數平滑法與一般線性迴歸。本研究建議，可將時間或環境等變動因素納入研究範圍內，進行探討，並與其他預測方法相互運用進而改善提高灰色預測模型的準確度。

關鍵字

糖果產業；灰色預測模型；平均絕對百分比誤差

並列摘要

The purpose of this study is to use Grey theory for establishment of the sales forecasting model, understanding the sales status of the candy industry in Taiwan, and it is used to be the basis of sales forecasting model. The research range is the sales volume of Taiwan's candy from 2000 year to 2011 year, using GM(1,1) model, four types of convertibility and Exponential Smoothing to forecast also analyze. At last using MAPE to measure the accuracy of model, moreover, finding the best forecasting method. The research results showed that GM(1,1) model is better than Exponential Smoothing and General Linear Regression for accuracy. This study suggests that variation factors such as time or environment can add to research range. Besides, this research can use other forecasting methods to promote accuracy of Grey forecasting model.

並列關鍵字

Candy Industry ； Grey Forecasting Model ； MAPE

參考文獻

Lewis, E.B., 1982. Control of Body Segment differentiation in Drosophila by the Bithorax Gene Complex, Embryonic Development, Part A: Genetics Aspects,Edited by Burger, M. M. and R. Weber. Alan R. Liss, New York, pp. 269-288.

于宗先，1972，「經濟預測」，中央研究院經濟研究所。