學術產出-學位論文

文章檢視/開啟

書目匯出

Google ScholarTM

政大圖書館

引文資訊

TAIR相關學術產出

題名 利用類神經網路估算國內電子股投資風險值績效
作者 高世儒
貢獻者 蔡瑞煌<br>林修葳
高世儒
關鍵詞 類神經網路
風險值
Artificial Neural Network
Value at Risk
日期 1998
上傳時間 18-九月-2009 19:32:06 (UTC+8)
摘要 本研究首次提出以未來臨界報酬率為輸出變數,利用兩種類神經網路(Artificial Neural Network)估算國內電子股代表樣本報酬率的風險值(Value at Risk , VaR)。在研究設計上考慮到使用不同期長來計算自變項所帶來的影響而產生兩種預測方法。本研究並以回顧檢定(Backtesting )檢討藉由臨界值報酬率作為類神經估計法與一般以變異數/共變數法或蒙地卡羅模擬法所估算出VaR的差異。
     綜合本研究,在學術及實務上的貢獻有下列四點:
     1. 設計臨界報酬率作為估算VaR的方式,可以避免以往計算VaR時,報酬率分配主觀給定的問題。
     2. 相關研究過去並未同時涉及類神經網路與VaR,而本研究首次應用類神經網路估算VaR。
     3. 本文亦提出以多種不同的基本變數衡量期長來估算VaR,或可幫助界定差異的研究設計。
     4. 本研究使用類神經網路可能的一項限制是報酬率臨界值 的設計方式;而類神經網路可能勝出其它預測工具的理由可能是 (1)學習到隱性因子的特性 (2)預測方式為非線性 (3)毋須依賴常態或特定分配之假設。以往類神經網路研究在賽馬決定各工具優劣時,較少探究類神經勝出或落敗的理由,而這卻是本研究設計的焦點。
參考文獻 一、中文部份:
1. 蔡瑞煌 (1995),類神經網路概論,台北:三民書局。
2. 蔡瑞煌、邱奕德、劉曦敏,「應用理解神經網路系統於臺灣股價指數之分
析及預測」,經濟研究,三十四卷,二期,民國85年7月,頁171-200。
二、英文部份:
1. Ahn, Dong-Hyun; Boudoukh, Jacob; Richardson, Matthew and
Whitelaw, Robert F,” Optimal Risk Management Using
Options,” Journal of Finance, Feb,1999,pp. 359-375.
2. Alexander C.O. and C.T. Leigh. “On the Covariance Matrices
Used In Value at Risk Models,” The Journal of Jerivatives,
Spring, 1997, pp. 50-62.
3. Beder, T. S., “VAR: Seductive but Dangerous,” Financial
Analysts Journal, September/ October 1995, pp. 12-24.
4. Chiou, Y., Liu, S. and Tsaih, R., “Applying Reasoning
Neural Networks to the Analysis and Forecast of Taiwan`s
Stock Index Variation,” Taipei Economics Inquiry , 1996,
pp.171-200.
5. David J. Ginzl., “How to Establish a Comprehensive Risk-
Management Program,” Commercial Lending Review, Summer
1997, pp. 31-35.
6. Duffie, D., and J. Pan “An Overview of Value at Risk,” The
Journal of Derivatives, Spring, 1997, pp. 7-49.
7. Hull, J., and A. White. “Value at Risk When Daily Changes
in Market Variables Are Not Normally Distributed,” The
Journal of Derivatives, Spring, 1998, pp. 9-19.
8. Jorion, P., “Risk: Measuring the Risk in Value at Risk,”
Financial Analysts Journal, November/December 1996, pp. 47-
56.
9. Jorion, P., (1997) Value at Risk: The New Benchmark for
Controlling Market Risk Professional Publication, U.S.A :
Iriwin
10. Paul H. Kupiec, “Techniques for Verifying the Accuracy of
Risk Measurement Models,” The Journal of Derivatives
Winter, winter 1995,pp. 73-85.
11. Tsaih, R., “Reasoning Network Networks,” In Ellacott, S.,
J. Mason and Anderson, I. (Eds.), Mathematics of Neural
Networks: Models, Algorithms and Applications, Kluwer
Academic Publishers, London, 1997,pp. 366-371.
12. Tsaih,R.,Chen, W. and Lin,Y. , “Application of Reasoning
Neural Networks for Financial Swaps,” Journal of
Computational Intelligence in Finance, May 1998, pp.27-37.
13. Tsaih,R.,Hsu,Y., and Lai,C., “Forecasting S&P500 Stock
Index Futures with the Hybrid AI System,” Decision Support
Systems, Jun 1998,pp.161-174.
14. Venkat, Shyam; Malhotra, Satyan. , “Establishing a Value-
at-Risk Framework,” Mortgage Banking, Aug 1998,pp. 83-86.
15. Venkataraman, “Value at Risk for Normal Distribution―the
Use of Quasi Bayesian Estimation Techniques,” Economic
Perspectives,1997,pp.12
描述 碩士
國立政治大學
資訊管理研究所
86356006
87
資料來源 http://thesis.lib.nccu.edu.tw/record/#B2002001641
資料類型 thesis
dc.contributor.advisor 蔡瑞煌<br>林修葳zh_TW
dc.contributor.author (作者) 高世儒zh_TW
dc.creator (作者) 高世儒zh_TW
dc.date (日期) 1998en_US
dc.date.accessioned 18-九月-2009 19:32:06 (UTC+8)-
dc.date.available 18-九月-2009 19:32:06 (UTC+8)-
dc.date.issued (上傳時間) 18-九月-2009 19:32:06 (UTC+8)-
dc.identifier (其他 識別碼) B2002001641en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/36761-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 資訊管理研究所zh_TW
dc.description (描述) 86356006zh_TW
dc.description (描述) 87zh_TW
dc.description.abstract (摘要) 本研究首次提出以未來臨界報酬率為輸出變數,利用兩種類神經網路(Artificial Neural Network)估算國內電子股代表樣本報酬率的風險值(Value at Risk , VaR)。在研究設計上考慮到使用不同期長來計算自變項所帶來的影響而產生兩種預測方法。本研究並以回顧檢定(Backtesting )檢討藉由臨界值報酬率作為類神經估計法與一般以變異數/共變數法或蒙地卡羅模擬法所估算出VaR的差異。
     綜合本研究,在學術及實務上的貢獻有下列四點:
     1. 設計臨界報酬率作為估算VaR的方式,可以避免以往計算VaR時,報酬率分配主觀給定的問題。
     2. 相關研究過去並未同時涉及類神經網路與VaR,而本研究首次應用類神經網路估算VaR。
     3. 本文亦提出以多種不同的基本變數衡量期長來估算VaR,或可幫助界定差異的研究設計。
     4. 本研究使用類神經網路可能的一項限制是報酬率臨界值 的設計方式;而類神經網路可能勝出其它預測工具的理由可能是 (1)學習到隱性因子的特性 (2)預測方式為非線性 (3)毋須依賴常態或特定分配之假設。以往類神經網路研究在賽馬決定各工具優劣時,較少探究類神經勝出或落敗的理由,而這卻是本研究設計的焦點。
zh_TW
dc.description.tableofcontents 壹、 緒論 1
     貳、 文獻探討 3
     一、 風險值(VAR)相關文獻 3
     二、 類神經網路相關文獻 9
     (一) BP神經網路 9
     (二) RN與RNBP神經網路 10
     參、 研究方法 12
     一、 兩種預測方法 15
     (一) 預測法一 15
     (二) 預測法二 17
     二、 VAR的驗證 19
     肆、 研究結果 22
     一、 三因子變異數分析的結果 22
     二、 預測法一的分析結果 27
     三、 預測法二的分析結果 29
     四、 變異數/共變數預測法與蒙地卡羅模擬法 32
     伍、 結論 34
     一、 研究結論 34
     (一) VaR估算方法的比較 34
     (二) BP神經網路與RNBP網路系統的差異 38
     (三) 兩種預測方法的差異 40
     二、 研究限制與未來研究方向 42
     文獻探討…………………………………………………………………………………………………..45
     附錄………………………………………………………………………………………...……………...47
zh_TW
dc.language.iso en_US-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#B2002001641en_US
dc.subject (關鍵詞) 類神經網路zh_TW
dc.subject (關鍵詞) 風險值zh_TW
dc.subject (關鍵詞) Artificial Neural Networken_US
dc.subject (關鍵詞) Value at Risken_US
dc.title (題名) 利用類神經網路估算國內電子股投資風險值績效zh_TW
dc.type (資料類型) thesisen
dc.relation.reference (參考文獻) 一、中文部份:zh_TW
dc.relation.reference (參考文獻) 1. 蔡瑞煌 (1995),類神經網路概論,台北:三民書局。zh_TW
dc.relation.reference (參考文獻) 2. 蔡瑞煌、邱奕德、劉曦敏,「應用理解神經網路系統於臺灣股價指數之分zh_TW
dc.relation.reference (參考文獻) 析及預測」,經濟研究,三十四卷,二期,民國85年7月,頁171-200。zh_TW
dc.relation.reference (參考文獻) 二、英文部份:zh_TW
dc.relation.reference (參考文獻) 1. Ahn, Dong-Hyun; Boudoukh, Jacob; Richardson, Matthew andzh_TW
dc.relation.reference (參考文獻) Whitelaw, Robert F,” Optimal Risk Management Usingzh_TW
dc.relation.reference (參考文獻) Options,” Journal of Finance, Feb,1999,pp. 359-375.zh_TW
dc.relation.reference (參考文獻) 2. Alexander C.O. and C.T. Leigh. “On the Covariance Matriceszh_TW
dc.relation.reference (參考文獻) Used In Value at Risk Models,” The Journal of Jerivatives,zh_TW
dc.relation.reference (參考文獻) Spring, 1997, pp. 50-62.zh_TW
dc.relation.reference (參考文獻) 3. Beder, T. S., “VAR: Seductive but Dangerous,” Financialzh_TW
dc.relation.reference (參考文獻) Analysts Journal, September/ October 1995, pp. 12-24.zh_TW
dc.relation.reference (參考文獻) 4. Chiou, Y., Liu, S. and Tsaih, R., “Applying Reasoningzh_TW
dc.relation.reference (參考文獻) Neural Networks to the Analysis and Forecast of Taiwan`szh_TW
dc.relation.reference (參考文獻) Stock Index Variation,” Taipei Economics Inquiry , 1996,zh_TW
dc.relation.reference (參考文獻) pp.171-200.zh_TW
dc.relation.reference (參考文獻) 5. David J. Ginzl., “How to Establish a Comprehensive Risk-zh_TW
dc.relation.reference (參考文獻) Management Program,” Commercial Lending Review, Summerzh_TW
dc.relation.reference (參考文獻) 1997, pp. 31-35.zh_TW
dc.relation.reference (參考文獻) 6. Duffie, D., and J. Pan “An Overview of Value at Risk,” Thezh_TW
dc.relation.reference (參考文獻) Journal of Derivatives, Spring, 1997, pp. 7-49.zh_TW
dc.relation.reference (參考文獻) 7. Hull, J., and A. White. “Value at Risk When Daily Changeszh_TW
dc.relation.reference (參考文獻) in Market Variables Are Not Normally Distributed,” Thezh_TW
dc.relation.reference (參考文獻) Journal of Derivatives, Spring, 1998, pp. 9-19.zh_TW
dc.relation.reference (參考文獻) 8. Jorion, P., “Risk: Measuring the Risk in Value at Risk,”zh_TW
dc.relation.reference (參考文獻) Financial Analysts Journal, November/December 1996, pp. 47-zh_TW
dc.relation.reference (參考文獻) 56.zh_TW
dc.relation.reference (參考文獻) 9. Jorion, P., (1997) Value at Risk: The New Benchmark forzh_TW
dc.relation.reference (參考文獻) Controlling Market Risk Professional Publication, U.S.A :zh_TW
dc.relation.reference (參考文獻) Iriwinzh_TW
dc.relation.reference (參考文獻) 10. Paul H. Kupiec, “Techniques for Verifying the Accuracy ofzh_TW
dc.relation.reference (參考文獻) Risk Measurement Models,” The Journal of Derivativeszh_TW
dc.relation.reference (參考文獻) Winter, winter 1995,pp. 73-85.zh_TW
dc.relation.reference (參考文獻) 11. Tsaih, R., “Reasoning Network Networks,” In Ellacott, S.,zh_TW
dc.relation.reference (參考文獻) J. Mason and Anderson, I. (Eds.), Mathematics of Neuralzh_TW
dc.relation.reference (參考文獻) Networks: Models, Algorithms and Applications, Kluwerzh_TW
dc.relation.reference (參考文獻) Academic Publishers, London, 1997,pp. 366-371.zh_TW
dc.relation.reference (參考文獻) 12. Tsaih,R.,Chen, W. and Lin,Y. , “Application of Reasoningzh_TW
dc.relation.reference (參考文獻) Neural Networks for Financial Swaps,” Journal ofzh_TW
dc.relation.reference (參考文獻) Computational Intelligence in Finance, May 1998, pp.27-37.zh_TW
dc.relation.reference (參考文獻) 13. Tsaih,R.,Hsu,Y., and Lai,C., “Forecasting S&P500 Stockzh_TW
dc.relation.reference (參考文獻) Index Futures with the Hybrid AI System,” Decision Supportzh_TW
dc.relation.reference (參考文獻) Systems, Jun 1998,pp.161-174.zh_TW
dc.relation.reference (參考文獻) 14. Venkat, Shyam; Malhotra, Satyan. , “Establishing a Value-zh_TW
dc.relation.reference (參考文獻) at-Risk Framework,” Mortgage Banking, Aug 1998,pp. 83-86.zh_TW
dc.relation.reference (參考文獻) 15. Venkataraman, “Value at Risk for Normal Distribution―thezh_TW
dc.relation.reference (參考文獻) Use of Quasi Bayesian Estimation Techniques,” Economiczh_TW
dc.relation.reference (參考文獻) Perspectives,1997,pp.12zh_TW