本論文提出以通用迴歸神經網路(General Regression Neural Network)建構一需求模式，並預測電子資訊產品之未來市場需求量。GRNN是從PNN(Probability Neural Network，機率類神經網路)所演變而來，主要應用在預測及控制上，可用來建立連續變數之函數關係，無論迴歸問題為線性或非線性均可用GRNN來解決。單變量時間序列、指數平滑法、迴歸模式是較常被使用在需求預測的方法。本論文則採用GRNN之模式，並與傳統之預測模式及整合模式比較，分別找出各預測模式最佳的參數組合，以探討GRNN是否比其他四種預測方法快且準確性高；本論文針對主機板銷售量進行實證研究，以經濟部統計處出版的工業生產統計初步速報作為研究驗證的研究資料。預測結果顯示，整合預測與GRNN之結果並無顯著差異，但是整合預測在進行學習時，花費較多時間，而GRNN學習及預測速度快，且比其餘的三種模式均來得佳，故GRNN在需求預測上有很好的表現。

In this thesis, we implement General Regression Neural Network into a demand model to forecast the demand of long-term and middle-term electronic products. The GRNN is a evolution from Probability Neural Network (PNN) and applied in control and forecasting problems for finding the relationship between continuous variables. One of the merits of GRNN is that GRNN can fit either linear or non-linear regression lines without extra efforts. Conventional statistical methods to forecast demands are unvariate time series, exponential smoothing method and regression analysis. In the research, performance evaluation of GRNN is studied and the comparisons among unvariate time series, exponential smoothing method, regression analysis and a weighted combination of several forecasting models are conducted. Mean absolute percent error (MAPE) and mean absolute deviations (MAD) are two performance indices used in the research. Studies show that both MAPE and MAD of the weighted linear combination models are smaller than any other models except GRNN model. There is no significant difference of MAPE and MAD between GRNN and weighted linear combination models. From time consumption viewpoint, GRNN is much shorter than the weighted linear combination models, thus the proposed GRNN model performs well in demand forecasting.

目 錄
中文摘要 ….………………………………………………………………….. i
英文摘要 ……………………………………………………………………. ii
致謝 …………………………………………………………………………….. iii
目錄 …………………………………………………………………………….. iv
表目錄 ……………………………………………………………………………. vi
圖目錄 ……………………………………………………………………….… vii
一、 緒論………………………………………………………………………... 1
1.1 研究動機與目的……………………………………………………… 1
1.2 研究架構…….……………………………………………………… 1
二、文獻探討…………………………………….………………………………... 3
2.1 GRNN之相關文獻探討………………………..…………..………… 3
2.2 整合(混合)預測模式之文獻探討…………………..……..………… 4
三、理論模式的簡介……………………………………………………………... 7
3.1 GRNN簡介……………………………………………………… 7
3.1.1 GRNN的理論背景………………..…………………………… 7
3.1.2 決定GRNN的平滑參數…………..…………………………… 9
3.1.3 GRNN的實行步驟……………..……………………………… 10
3.1.4 GRNN與其他類神經網路之不同……………………..……… 10
3.2 單變量時間序列模式簡介…………………………………………… 11
3.3 指數平滑法簡介……………………………………………………… 13
3.4 迴歸模式簡介………………………………………………………… 15
3.5 整合模式簡介………………………………………………………… 16
3.6 評估指標……………………………..……………………………… 17
四、問題定義與模式建構…………………………………………..…………….. 18
4.1 影響主機板之銷售量因素之探討…………………………………… 18
4.2 各預測模式在需求模式上的建構…………………………………… 18
4.2.1 GRNN在需求模式上的建構…………...…………………… 18
4.2.2 單變量時間序列在需求模式上的建構……………..………… 20
4.2.3 指數平滑法在需求模式上的建構………...………...………… 20
4.2.4 迴歸模式在需求模式上的建構…………………….…….…… 21
4.2.5 整合預測在需求模式上的建構……………………..………… 21
五、實例分析與討論………………………………………………………………. 22
5.1 通用迴歸類神經網路……………..………………………………….. 22
5.2 單變量時間序列模式…….…...……………………………………… 25
5.3 指數平滑法…………………………………………………………… 26
5.4 迴歸模式…………………………..………………………………….. 28
5.5 整合模式……………………………………………………………… 30
5.6 模式預測能力之比較與預測誤差分析……………………………… 31
六、結論………………………………………………………………………... 33
參考文獻………………………………………………………………………... 34

參考文獻
1. 吳宗正，回歸分析，三民書局，台北(1993)。
2. 林茂文，時間數列分析與預測，華泰書局，台北(1992)。
3. 林聰明，吳水丕，指數平滑法之選擇與應用，華泰書局，台北(1971)。
4. 時巧煒，「來華觀光旅客需求預測模式建立之研究」，國立政治大學統計研究所碩士論文，台北(1994) 。
5. 黃傳欽，「台灣地區白肉雞產地價格組合預測方法之研究」，嘉義農專學報，第32期，1993，pp. 121-142。
6. 葉怡成，類神經網路模式應用與實作，儒林出版社，台北(1998)。
7. 葉怡成，應用類神經網路，儒林出版社，台北(1997)。
8. 劉亭宜，「GRNN在晶圓製造裡良率模式之建構與分析」，私立元智大學工業工程研究所碩士論文，中壢(2000) 。
9. Disorntetiwat, P. and C. H. Dagli, “Simple Ensemble-averaging Model Based on Reneralized Regression Neural Network in Financial Forecasting Problems,” Adaptive Systems for Signal Processing, Communications and Control Symposium 2000 IEEE AS-SPCC, 2000, pp. 477—480.
10. Dong, J., “Reseach on the Method of Non-linear Combining Forecasting Based on Fuzzy Neural System,” Proceedings. of the 3th World Congress on Intelligent and Automation, 2000, pp. 899-903.
11. Frost, F. and V. Karri, “Performance Comparison of BP and GRNN Models of the Neural Network Paradigm Using a Practical Industrial Application,” Proceedings. of ICONIP 1999 6th International Conference on Neural Information, Vol. 3, 1999, pp. 1069-1074.
12. James, T. L., O. R. Jens and S. Brian, “A Hybrid Econometric-neural Network Modeling Approach for Sales Forecasting,” Tourism Management, Vol. 27, 1996, pp. 175-192.
13. Kim, Y. S, B. J. Yum and M. Kim, “A Hybrid Model of Partial Squares and Artificial Neural Network for Analyzing Process Monitoring Data,” 2000 IEEE, 2001, pp. 2292-2297.
14. Lee, K. C., I. Han and Y. Kwon, “Hybrid Neural Network Model for Bankruptcy Predictions,” Decision Support Systems, Vol. 18, 1996, pp. 63-72
15. Leung, M. T., A. S. Chen and H. Daouk, “Forecasting Exchange Rates using General Regression Neural Networks,” Computers & Operations Research, Vol. 27, 2000, pp.1093-1110.
16. Masters, T., W. H. Land, Jr., J. Y. Lo and D. W. McKee, “Application of Evolutionary Computation and Neural Network Hybrids for Breast Cancer Classification Using Mammogram and History Data,” Proceedings of the 2001 IEEE Congress on Evolutionary Computation, Vol. 2, 2001, pp. 1147-1154.
17. Patton, J. B. and J. Ilic, “Identification of Static Distribution Load Parameters Using General Regression Neural Networks,” Proceedings of the 36th Midwest Symposium on Circuits and Systems, Vol. 2, 1993, pp.1023-1026.
18. Shi, M. S., D. X. Li and B. Liu,“Improving the Accuracy of Non-linear Combined Forecasting Using Neural Networks,”Expert Systems with Applications, Vol. 16, 1999, pp. 49-54.
19. Specht, D. F.,“A General Regression Neural Network,”IEEE Transactions on Neural Networks, 1991, pp. 568-576.
20. Yu, S. and L. Guan, “Feature Selection Using General Regression Neural networks for the Automatic Detection of Clustered Microcalcifications,” Proceedings of the 1999 IEEE International Conference on Acoustics Speech and Signal, Vol. 2, 1999, pp. 1101—1104.