臺灣博碩士論文加值系統

需求預測為存貨管理面最重要的一環，因為其預測結果將會成為生產計畫和存貨水準的依據。企業管理人員為了決定適當的生產計畫和安全存貨量必須面對消費者多變的消費心態，能否準確的預測未來需求會影響到決策的品質。降低預測的誤差能增加存貨空間和流動現金，以達到降低存貨管理的成本。
在眾多的預測法中，誰優誰劣已經不是大家所關心的議題，因為大家都已体會到一個事實，那就是世上沒有一種絕對比較好的預測法。混合預測法並不是在此一領域出現的一項革新方法，而是一項提昇現存的預測法精確度的技巧。
本研究以十六種需求型態為環境，比較三種業界常用的預測法和合併法的準確度，並以實驗証明合併法的精確度不管在何種需求型態下都高出三種傳統預測法有10%以上。

Demand forecasting is one of the most crucial issues of inventory management since production planning and inventory management are based on demand forecasts. To satisfy the random demands of customers, managers need to decide appropriate production planning and inventory levels. Reducing the demand forecasting error can free up capital and inventory space, which leads to reduction in the cost of inventory management.
There exists no unique best forecasting method for different demand models. Everyone has already known that there is no best forecasting method fitting all situations in the world. Combining forecasting method use historical data to select the most promising method in each of the three types of forecasting method. Then a quadratic program is solved for the appropriate weights on each method such that the minimum mean square errors is achieved.
Using mean squared errors as a comparison criterion, this study compares three common types of forecasting methods with the combining-method in sixteen different situations. Empirical show that the combining-method is outperforms the existing methods by more than ten percent in the mean squared errors..

目錄
中文摘要 ---------------------------------------------------Ⅰ
英文摘要 ---------------------------------------------------Ⅱ
誌謝 ---------------------------------------------------Ⅲ
目錄 -------------------------------------------------Ⅳ
圖表索引 ---------------------------------------------------Ⅴ
第一章 緒 論
1.1動機與目地 --------------------------------------------1
第二章 相關文獻及理論探討
2.1預測簡介 --------------------------------------------3
2.2指數平滑法
2.2.1 簡單指數平滑法 -----------------------------------4
2.2.1.1 簡單指數平滑法的期望值 --------------------------6
2.2.2線性指數平滑法 -----------------------------------7
2.2.2.1線性指數平滑法的期望值 --------------------------7
2.2.3應變指數平滑法 -----------------------------------9
2.3時間序列分析 -------------------------------------------11
2.3.1 ARMA模型簡介 ----------------------------------11
2.3.1.1 AR(p)模型定義------------------------------------11
2.3.1.2 MA(q)模型定義 ----------------------------------12
2.3.1.3 ARMA(p,q)模型定義 -------------------------12
2.3.2 ARMA各參數估計量的選擇 -------------------------12
2.3.2.1 AR(p)的參數估計 -------------------------13
2.3.2.2 MA(q)的參數估計 -------------------------13
2.4時間迴歸分析法 -------------------------------------------15
2.4.1 線性迴歸法 -------------------------------------------15
2.4.2 多項式迴歸法 ----------------------------------15
2.4.3 轉換變數迴歸法 ----------------------------------18
第三章 實驗之方法與步驟
3.1模型導入 -------------------------------------------21
3.2對於趨勢因子和季節因子的處理 -------------------------22
第四章 實驗結果分析 -------------------------24
第五章 結論與建議 -------------------------30
參考資料 -------------------------------------------31
附錄 ----------------------------------------------------32
作者簡介 ----------------------------------------------------66
授權書 -------------------------------------------67

參考文獻
中文文獻：
林聰明、吳水丕，指數平滑法之選擇與應用，第153~第155頁，華泰書局，民國七十年。
英文文獻：
Chi Kin Chan , Brian G. Kingsman , H.Wong , “The value of combining forecasts in inventory management --- a case study in banking.” EJOR 117,199-210(1999).
Bates,J.M., Granger, C.W.J., “The combination of forecasts.” Operational Research Quarterly 20,451-468(1969).
Granger, C.W.j., Ramanathan,R., “Improved methods of combining forecasts.” Journal of Forecasting 3 , 197-204(1984).
Clemen, R.T., “Combining forecasts: A review and annotated bibliography.” International Journal of Forecasting 5 , 559-583(1989).
Newbold,P.,Granger,C.W.J., “Experience with forecasting univariate time series and the combination of forecasts(with discussion) .” Journal of the Royal Statistical Society Series A 137,131-149(1974).
Winkler, R.L., Makridakis, S., “The combination of forecasts .” Journal of the Royal Statistical Society, Series A,146,150-157(1983).
Holt,C.C., “Forecasting Seasonal and Trends by Exponentially Weighted Moving Averages.” Carnegie Institute of Technology, Pittsburgh,Pa(1975).
Trigg, D W. , and D. H. Leach, “Exponential Smoothing with an Adaptive Response Rate.” Operational research Quarterly, 18 ,53-59(1967).
Peter J. Brockwell , Richard A. Davis, “Introduction to Time Series and Forecasting .”Springer Texts in Statistics (1996).
Kruskal, W., and W. Wallis “Use of Ranks In One-Criterion Variance Analysis,”Journal of the American Statistical Association 67 : 401-412(1952)