臺灣博碩士論文加值系統

近幾年來，貴金屬材料中的白銀具備絕佳多功能材料特性，使得製造工業需求大增，然而白銀卻處於高度投資投機性操作市場中，造成白銀價格大幅度攀升且漲勢不斷，致使相關工業需求者面臨沈重的採購成本壓力，因此若能準確預測白銀價格將可有效舒緩需求者背負的原物料採購壓力。
本研究乃應用倒傳遞類神經網路(Back-Propagation Neural Networks,BPNN)預測每日白銀價格，這可使相關工業需求者實際應用於每日工作中。模型輸入變數包含直接性及投資需求因素，待執行完結合不同網路架構及學習率的測試後，即提出一較佳BPNN模型進行後續實驗，並進一步驗證加入時間序列變數以調整BPNN模型輸入的結果。
經實證研究顯示，最佳網路架構設定值為：一層隱藏層、七個隱藏層處理單元數、十二個輸入變數及一個輸出變數。投資者需求為一影響白銀價格預測準確度的關鍵變數，其他主要變數包含直接性因素中的相同貴金屬群及共生副產品群價格。另為測試白銀價格預測準確度，以連續30日為一驗證期間，得到平均預測準確率為98.7%。這證明本研究所建構的類神經網路模型可有效預測白銀之價格。

Among those of precious materials, silver has many superior characteristics that the silver demand in industry boosts in recent years. Moreover, as silver is in a highly speculative market of investment, the price goes up and variates substantially at the same time. The industry is now confronting with high pressure of costs for purchasing silver, which makes accuracy prediction of silver price becomes important for companies in demand for silver.
The research is to apply backpropagation artificial neural networks (BPNN) to predict daily silver price that practitioners can implement in their daily work. We include direct-related and investment factors as the input to the BPNN. After performing different combinations of network structure and learning rate, we propose a reasonably good BPNN model for the experiment. We further experiment with time series variables to adjust the inputs of the BPNN model.
The resultant network structure has one hidden layer with 7 nodes, 12 input nodes and one output node. Experiments also show that the demand amount of investors is one of the key variables affecting the accuracy of silver price prediction. Other main variables include the price of equivalent precious metals and symbiosis byproducts in the immediateness factor. We tested the accuracy of silver price prediction for a period of consecutive 30 days and found an average accuracy of 98.7%. This demonstrates the effectiveness of our BPNN model in silver price prediction.

目錄 iii
表目錄 v
圖目錄 vi
壹、序論 1
1.1 研究背景與動機 1
1.2 研究目的 3
1.3 研究架構及流程 3
貳、文獻探討 4
2.1 類神經網路簡介 4
2.1.1類神經網路定義及概念 5
2.1.2基本架構及網路模式 6
2.2 類神經網路預測模型應用 10
2.3 貴金屬材料市場相關性研究 14
2.4 影響銀價波動之因素 17
2.4.1需求面因素 17
2.4.2供給面因素 19
2.4.3直接性因素 25
2.4.4投資需求因素 26
參、研究方法 30
3.1 倒傳遞類神經網路 30
3.1.1基本原理及網路架構 30
3.1.2網路學習演算 31
3.1.3參數設定 33
3.2 研究資料來源限定 34
3.3 變數資料前處理 35
3.4 網路模式架構建立 36
3.5 網路模型訓練及學習 37
肆、實證研究與分析 38
4.1 隱藏層處理單元數前測結果 38
4.2 第一階段不同學習速率實證結果 42
4.3 第二階段改變學習循環次數實證結果 44
4.4 第三階段預測模型驗證結果 48
伍、結論與建議 57
陸、參考文獻 60
中文文獻 60
英文文獻 61
網路文獻 65

中文文獻
1.世界白銀年鑑（2006），英國黃金礦業服務有限公司受世界白銀協會委託編制，孫鳳民總編，中國珠寶玉石首飾行業協會與世界白銀協會聯合出版，17-94。
2.世界白銀年鑑（2007），英國黃金礦業服務有限公司受世界白銀協會委託編制，王耀總編，中國商業聯合會與世界白銀協會聯合出版，7-95。
3.李宏志、賴秀卿（2001），黃金與白銀期貨間套利機會之探討－根據每日最後小時資料，成功大學學報，36，65-90。
4.李沃牆、李建信（2003），台指選擇權之評價－ANN與GANN模型之績效比較，真理財經學報，8，25-50。
5.余尚武、黃雅蘭（2003），台灣股價指數期貨套利之研究－類神經網路與灰色理論之應用，電子商務學報，5，87-115。
6.吳宗正、溫敏杰、侯惠月（2001），類神經網路及統計方法在台股指數期貨預測研究之比較，成功大學學報，36，91-109。
7.林美蓮（2001），高頻率股市報酬波動性之ANN－GARCH MODEL，交通大學經營管理研究所碩士論文。
8.柯淑玲（2000），運用類神經網路於台股認購權證評價模式之實證研究，義守大學管理研究所碩士論文。
9.黃政仁（1996），貴金屬市場中黃金與白銀期貨間互動性及效率性之探討－根據每小時資料，國立成功大學會計學研究所碩士論文。
10.黃華山、邱一薰（2005），類神經網路預測台灣50股價指數之研究，資訊科技與社會學報，5，19-42。
11.陳安斌、許育嘉（2004），整合小波轉換與神經網路於金融投資決策時間序列預測之研究，資訊管理學報，11（1），139-165。
12.葉怡成（2004），類神經網路模式應用與實作，台北：儒林圖書。
英文文獻
1.Agnon, Y., Golan, A., & Shearer, M. (1999). Nonparametric, nonlinear, short-term forecasting: theory and evidence for nonlinearities in the commodity markets. Economics Letters, 65, 293-299.
2.Aguirre, A., & Aguirre, L. A. (2000). Time series analysis of monthly beef cattle prices with nonlinear autoregressive models. Applied Economics, 32(3), 265-275.
3.Ailawadi, K. L., & Neslin, S. A. (1998). The effect of promotion on consumption: buying more and consuming it faster. Journal of Marketing Research, 35(3), 390-398.
4.Amilon, H. (2003). A neural network versus black-scholes: a comparison of pricing and hedging performance. Journal of Forecasting, 22(4), 317-335.
5.Brooks, C. (1998). Predicting stock index volatility: can market volume help? Journal of Forecasting, 17(1), 59-80.
6.Cai, J., Cheung, Y. L., & Wong, M. C. S. (2001). What moves the gold market? Journal of Futures Markets, 21(3), 257-278.
7.Chow, Y. F. (2001). Arbitrage, risk premium, and cointegration tests of the efficiency of futures markets. Journal of Business Finance and Accounting, 28(5/6), 693-713.
8.Christie-David, R., Chaudhry, M., & Koch, T. W. (2000). Do macroeconomics news releases affect gold and silver prices? Journal of Economics and Business, 52(5), 405-421.
9.Ciner, C. (2001). On the long run relationship between gold and silver prices, A note. Global Finance Journal , 12(2), 299-303.
10.Cornell, B., & French, K. (1986). Commodity own rates, real interest rates, and money supply announcements. Journal of Monetary Economics, 18(1), 3-20.
11.Dhillon, U. S., Lasser, D. J., & Watanabe, T. (1997). Volatility, information, and double versus walrasian auction pricing in US and Japanese futures markets. Journal of Banking & Finance, 21(7), 1045-1061.
12.Donaldson, G. R., & Kamstra, M. (1996). Forecast combining with neural networks. Journal of Forecasting, 15(1), 49-61.
13.Ederington, L., & Lee, J. (1993). How markets process information: news releases and volatility. Journal of Finance, 48(4), 1161-1191.
14.Einhorn, C. S. (1995). What’s sparking silver? Barron’s May 15, 75(20): MW 14(1).
15.Erdem, T. (1996). A dynamic analysis of market structure based on panel data. Marketing Science, 15(4), 359-378.
16.Escribano, A., & Granger, C. W. J. (1998). Investigating the relationship between gold and silver prices. Journal of Forecasting, 17(2), 81-107.
17.Fineberg, S. (1996). Silver analysis sees stronger mart. American Metal Market May 16, 104.
18.Frankel, J., & Hardouvelis, G. A. (1985). Commodity prices, money surprises, and Fed credibility. Journal of Money, Credit and Banking, 17(4), 425-438.
19.Frees, E. W., & Miller, T. W. (2004). Sales forecasting using longitudinal data models. International Journal of Forecasting, 20(1), 99-114.
20.Hamid, S. A., & Zahid, I. (2002). Using neural networks for forecasting volatility of S&P500 index futures prices. Journal of Business Research, 5881, 1-10.
21.Hibon, M., & Evgeniou, T. (2005). A simple procedure for reliability of repairable systems. International Journal of Forecasting, 21, 15-24.
22.Hill, S. R., Moore, N. H., & Pruitt, S. W. (1991). Gold fusion-hot metal: An analysis of the metals futures market reactions to the cold fusion announcement. Journal of Future Mark, 11(3), 385-397.
23.Hill, T., O’Connor, M., & Remus, W. (1996). Neural networks models for time series forecasts. Manage Science, 42(7), 1082-1092.
24.Hsieh, K. L. (2001). Process improvement in the presence of qualitative response by combing fuzzy sets and neural networks. Integrated Manufacturing Systems, 12(6-7), 449-462.
25.Ma, C. K. (1985). Spreading between the gold and silver markets：is there a parity? The Journal of Futures Markets, 5(4), 579-594.
26.Ma, C. K., & Soenen, L. A. (1988). Arbitrage opportunities in metal futures markets. The Journal of Futures Markets, 8(2), 199-209.
27.Mahdavi, S., & Zhou, S. (1997). Gold and commodity prices as leading indicators of inflation: tests of long-run relationship and predictive performance. Journal of Economics and Business, 49(5), 475-489.
28.Mcloone, S., Brown, M. D., Irwin, G., & Lightbody, A. (1998). A hybrid linear/nonlinear training algorithm for feedforward neural networks. IEEE Transactions on Neural Networks, 9(4), 669-684.
29.Mirmirani, S., & Li, H. C. (2004). Gold price, neural networks and genetic algorithm. Computational Economics, 23(2), 193-200.
30.Nakhjavani, O. B., & Ghoreishi, M. (2006). Multi criteria optimization of laser percussion drilling process using artificial neural network model combined with genetic algorithm. Materials and Manufacturing Process, 21(1), 11-18.
31.Narendra, K., & Parthasarathy, K. (1990). Identification and control of dynamical systems using neural networks. IEEE Transactions on Neural Networks, 1(1), 4-27.
32.Parisi, A., Parisi, F., & Diaz, D. (2007). Forecasting gold price changes: rolling and recursive neural network models. Journal of Multinational Financial Management, 1-14.
33.Ramazan, G. (1996). Non-linear prediction of security returns with moving average rules. Journal of Forecasting, 15(3), 165-174.
34.Sexton, R. S., Alidaee, B., Dorsey, R. E., & Johnson, J. D. (1998). Toward global optimization of neural networks: a comparison of the genetic algorithms and backpropagation. Decision Support Systems, 22(2), 171-185.
35.Shaikh, A., & Zahid, I. (2004). Using neural networks for forecasting volatility of S&P 500 index futures prices. Journal of Business Research, 57(10), 1116-1125.
36.Timmermann, A., & Granger, C. W. J. (2004). Efficient market hypothesis and forecasting. International Journal of Forecasting, 20(1), 15-27.
37.Tong, L. I., & Liang, Y. H. (2005). Forecasting field failure data for repairable systems using neural networks and SARIMA model. International Journal of Quality and Reliability Management, 22(4), 410-420.
38.Urich, T. J. (2000). Models of fluctuation in metal futures prices. Journal of Futures Markets, 20(3), 219-241.
39.Varela, O. (1999). Futures and realized cash or settle prices for gold, silver, and copper. Review of Financial Economics, 8(2), 121-138.
40.Wahab, M., Cohn, R., & Lashgari, M. (1994). The gold-silver spread: integration, cointegration, predictability and ex-ante arbitrage. Journal of Futures Markets, 14(6), 707-756.
41.Webby, R., & O’Connor, M. (1996). Judgmental and statistical time-series forecasting: a review of the literature. International Journal of Forecasting, 12(1), 91-118.
42.Xu, X. E., & Fung, H. G. (2005). Cross-market linkages between U.S. and Japanese precious metals futures trading. International Financial Markets, Institutions and Money, 15(2), 107-124.
43.Zhang, G., Patuwo, B. E., & Hu, M. Y. (1998). Forecasting with artificial networks: the state of the art. International Journal Forecasting, 14(1), 35-62.
網路文獻
1.中國白銀網（2007），上海華通鉑銀講評，http://www.ex-silver.com/。
2.世界日報（2006），取自http://blog.xuite.net/jsksonic/sycee/10235867。
3.Kitco網站（2008），http://www.kitco.com。