English  |  正體中文  |  简体中文  |  Post-Print筆數 : 27 |  Items with full text/Total items : 113311/144292 (79%)
Visitors : 50940508      Online Users : 976
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    政大機構典藏 > 商學院 > 資訊管理學系 > 學位論文 >  Item 140.119/76425
    Please use this identifier to cite or link to this item: https://nccur.lib.nccu.edu.tw/handle/140.119/76425


    Title: 運用資料探勘技術於台幣匯率趨勢預測之研究
    A Study of Applying Data Mining to Predict the Trend of TWD Exchange Rate
    Authors: 陳威宇
    Contributors: 楊建民教授
    陳威宇
    Keywords: 資料探勘
    倒傳遞類神經網路
    匯率預測
    灰色地帶
    Data Mining
    BP Neural Network
    Exchange Rate
    Grey Area
    Date: 2015
    Issue Date: 2015-07-13 11:07:42 (UTC+8)
    Abstract: 台灣地狹人稠,自然資源不足,台灣絕大數的資源必須仰賴國際貿易的補充,來維持經濟的發展,國際貿易可說是台灣經濟的命脈。在國貿中,匯率的變化更是深深影響每筆交易,對於政府與投資者都是不可不重視課題。過去對於匯率影響相關因素之研究許多,但欠缺一個綜合整體因素之研究,本研究就已整合過去文獻為出發點,延伸至對輸入資料進行優化,並利用滾動建模方式建立投資組合,提供投資者適當之進場時間與投資報酬率區間之參考依據
    本研究建立倒傳遞類神經網路模型,整理學者對於台幣匯率影響之因素如國際收支、外匯存底、主要貿易國匯率等共27項,以自2003年7月到2014年10月資料為模型輸入,預測以美元為標準的月平均台幣匯率漲跌,並比較過濾灰色地帶(Grey Area)之模型與原模型之預測能力差異,再以滾動建模方式觀察不同進場時間之平均獲利與投資報酬上下限值。
    結果發現,有過濾輸入值與輸出值之灰色地帶模型預測能力優於只過濾輸入值之灰色地帶模型,而預測能力在四者之中最差為未過濾之模型。而在投資報酬率部分,有過濾輸入輸出值之灰色地帶模型進場一到三年之平均報酬率介於5.02%~6.13%,獲利區間為2.8%~10.84%;只過濾輸入值之灰色地帶模型進場一到三年之平均報酬率介於4.39%~5.72%,獲利區間為1.17%~10.84%;未過濾灰色地帶之模型進場一到三年之平均報酬率介於3.86%~5.18%,獲利區間為-0.71%~10.54%,本研究之結果可在政府及投資人投資決策上給予具有參考性的指標。未來研究方向可加入利用文字探勘或情緒探勘來觀察政策對匯市所造成之影響或觀察金融海嘯前資料的變化來預測金融危機之發生。
    Taiwan is a small island with limited land but huge population, and the nature resources can’t afford to the economic development. Therefore, we have to rely on international trades to supplement our resources. However, exchange rate plays an important role in international trades, because it impacts each trade directly. For government and investors, exchange rate is an important topic they should notice.
    This study consolidated the scholars’ studies on the factors impacting the exchange rate to build a backward-propagation neural network model. We collected 27 variables (such as BoP, inflation rate, GDP…) as model’s input, and the prediction of exchange rate’s appreciation or depreciation is the output. The study compares four kinds of model’s accuracy, precision, recall and ROI. One is filter grey area data in input and output, and the second is only filter grey area data in input data, and the third is only filter grey area data in output result, and the last is no filter any data.
    The result shows that the first model is the best. Its average ROI of one year is between 5.02%~6.13%, and the profit range is between 2.8%~10.84%. This study suggests investors to collect past seven years data and filter the grey area data to build up the model to predict the next year’s exchange rate. Beside, when we saw the predict values in grey area, we should not invest. Follow this rule, this model will be helpful for investors when they need references to make exchange rate investment decisions. For future researches, I suggest that one way is to use text mining or motion mining to find out how the policies impact exchange rate market. The other way is to observe the change before economic crisis trying to predict the happening times for helping investors to avoid the economic crisis.
    Reference: 王允俊,匯率、金價與油價關係之研究,國立高雄應用科技大學金融資訊研究所碩士論文,2008
    邱世鈞,情緒指標對台股指數期貨報酬率之影響,中興大學財務金融學系所碩士論文,2009
    林柏君、吳中書,通膨與通縮之匯率轉嫁,台灣經濟預測與政策43:2,51-81,2013
    周香源,台灣、新加坡、南韓匯率影響因素的研究--Branson資產組合平衡模型應用,佛光大學經濟學系碩士論文,2005
    施向陽,匯率變動預測模型之研究,大葉大學事業經營研究所碩士論文,2001
    孫珮君,情緒指標對黃金期貨價格的影響,中央大學財務金融學系在職專班論文,2014
    陸寒寅,再議金本位制和30年代大危機:起因、擴散和復甦,世界經濟專刊2008年05期,2008
    陳裕菘、謝邦昌、李勝輝、陳郁婷,運用文字探勘與資料採礦技術建立匯率預測模型-以人民幣兌新台幣為例,「數據分析」2014年2月,9卷1期,頁133-146,2014
    陳益明,新台幣匯率與台灣貿易收支:時間數列迴歸模型的分析,國立空中大學商學系期刊論文,2001
    黃朝熙(2008),價格僵固能否解釋購買力平價之謎?跨國的實證研究,國科會計畫-ECON Projects 編號NSC94-2410-H007-006,2008
    黃欣華,影響台灣匯率因素之實證研究,台灣大學經濟學研究所碩士論文,2008
    黃惠琪,台美匯率之迴歸分析,國立中央大學統計學系碩士論文,2002
    張瓊文、張瑞芳,應用基因演算法與倒傳遞類神經網路於匯率預測模型之開發,嘉南學報,36,270-279,2010
    張伯群,中央銀行的匯率政策:金融風暴前後的跨國比較,中國文化大學經濟學系碩士論文,2012
    葉柏村,運用類神經網路預測匯率─以歐元為例,中原大學企業管理研究所碩士論文,2002
    葉怡成,應用類神經網路。台北市:儒林圖書,2004
    傅澤偉、丁裕家,匯率與總體經濟變數之連動性探討-台灣、美國、日本及韓國實證研究,第七屆兩岸產業發展與經營管理學術研討會,2008
    鄭漴瑋,應用灰關聯分析與類神經網路於預測人民幣匯率之研究,義守大學財務金融學系碩士論文,2014
    賴松鐘,外匯匯率與黃金價格長期互動關係之研究,政大金融學系財政研究所碩士論文,1994
    羅華強,類神經網路,新北市:高立圖書,2011
    施柏屹,倒傳遞類神經網路學習收斂之初步探討,中央大學機械工程研究所碩士論文,2000
    蘇木春、張孝德,機械學習類神經網路、模糊系統以及基因演算法則,台北市:全華圖書,1997
    Beal R. and Jackson T.,Neural Computing: An Introduction,Adam Hilger., 1990
    Freisleben B.,“Stock Market Prediction with Back Propagation Neural Networks”,Industrial and Engineering Application of Artificial Intelligence and Expert System. 5th International Conference, Paderborn,Germany June P.451-460., 1992
    Mirmirani, S. and Li, H. C., Glod Price, Neural Network and Genetic Algorithm, Computational Economics, 23(2), pp.193-200, 2004
    Refenes A. N., Zapranis A. and Francis G.,“Stock performance modeling using neural networks: a comparative study with regression models”, Neural Network Vol. 5, p.961-970., 1994
    Yoon, Y., and Swales, G., Predicting stock price performance : A neural network approach, The Mit Press, vol. 25,no. 11,pp. 39-44,1991
    Description: 碩士
    國立政治大學
    資訊管理研究所
    102356028
    103
    Source URI: http://thesis.lib.nccu.edu.tw/record/#G0102356028
    Data Type: thesis
    Appears in Collections:[資訊管理學系] 學位論文

    Files in This Item:

    File Description SizeFormat
    602801.pdf3616KbAdobe PDF2389View/Open


    All items in 政大典藏 are protected by copyright, with all rights reserved.


    社群 sharing

    著作權政策宣告 Copyright Announcement
    1.本網站之數位內容為國立政治大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
    The digital content of this website is part of National Chengchi University Institutional Repository. It provides free access to academic research and public education for non-commercial use. Please utilize it in a proper and reasonable manner and respect the rights of copyright owners. For commercial use, please obtain authorization from the copyright owner in advance.

    2.本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本網站維護人員(nccur@nccu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
    NCCU Institutional Repository is made to protect the interests of copyright owners. If you believe that any material on the website infringes copyright, please contact our staff(nccur@nccu.edu.tw). We will remove the work from the repository and investigate your claim.
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback