| dc.contributor.advisor | 林士貴<br>張宜武 | zh_TW |
| dc.contributor.advisor | Lin, Shih-Kuei<br>Chang, Yi-Wu | en_US |
| dc.contributor.author (Authors) | 許力夫 | zh_TW |
| dc.contributor.author (Authors) | Hsu, Li-Fu | en_US |
| dc.creator (作者) | 許力夫 | zh_TW |
| dc.creator (作者) | Hsu, Li-Fu | en_US |
| dc.date (日期) | 2018 | en_US |
| dc.date.accessioned | 4-Jan-2019 16:34:03 (UTC+8) | - |
| dc.date.available | 4-Jan-2019 16:34:03 (UTC+8) | - |
| dc.date.issued (上傳時間) | 4-Jan-2019 16:34:03 (UTC+8) | - |
| dc.identifier (Other Identifiers) | G1047510151 | en_US |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/121735 | - |
| dc.description (描述) | 碩士 | zh_TW |
| dc.description (描述) | 國立政治大學 | zh_TW |
| dc.description (描述) | 應用數學系 | zh_TW |
| dc.description (描述) | 1047510151 | zh_TW |
| dc.description.abstract (摘要) | 本研究利用2008/01/04至2018/09/28台灣加權指數進行分析與評估風險值之預測效果。研究成果與貢獻如下：一、延伸GARCH模型，假設即期波動度與非預期變動、歷史波動度呈非線性關係，透過類神經網路來捕捉更多非線性槓桿、波動叢集等效果。二、針對不同模型，分別利用最大概似法、反向傳播演算法等進行參數估計與訓練網路。實證結果顯示新模型對於波動度具有較好之配適能力。三、與變異數法、歷史模擬法、GARCH 模型比較，在95%信心水準下新模型所計算之風險值具有較低之穿透率。因此新模型所計算之風險值應可有效改善企業投資時所需提撥之準備金，主管機關亦可透過此模型來訂定投資人所需付出之保證金，皆可降低信用風險與穩定金融市場。本研究可提供上述產業評估風險時較為精準、客觀與較有效率之工具。 | zh_TW |
| dc.description.abstract (摘要) | We analyze the VaR prediction by using the TWII data from 2008/01/04 to 2018/09/28. The contribution and results are as following: First, to extend the GARCH model, we assume spot volatility, unexpected volatility and historical volatility have non-linear relationship. By training neural network, we capture more non-linear lever effects and cluster volatility effects. Second, compared with different VaR models, we use Max Likelihood method to estimate the parameters and Backpropagation to train the neural network. The results show that the new model fits the volatility better than others. Third, compare the new model with other methods, VaR values predicted by new model have lower ABLF values. Therefore, the VaR values evaluated by new model can improve the reserve fund when the enterprise invests. The financial authority also can set the security deposit by using new model. This study can provide the abovementioned industrial a precise and objective tool to evaluate the risk. | en_US |
| dc.description.tableofcontents | 目 錄摘 要 2ABSTRACT 3目 錄 4圖目錄 5表目錄 6第一章 緒論 7第一節 研究背景與動機 7第二節 研究目的 8第二章 文獻探討 9第一節 國內外傳統風險值模型比較之文獻 9第二節 類神經網路應用於風險值模型 11第三章 研究方法 13第一節 風險值介紹 13第二節 共變異數法(VARIANCE-COVARIANCE APPROACH) 14第三節 歷史模擬法(HISTORICAL SIMULATION) 14第四節 GARCH (1,1)模型 16第五節 GARCH(1,1)概念結合神經網路 17第六節 風險值模型準確性評估方法 21第四章 實證資料 24第一節 資料期間 24第二節 日資料敘述性統計 24第三節 模型建構：移動視窗法 25第四節 神經網路超參數設定 25第五章 實證結果 29第六章 結論 32參考文獻 34 | zh_TW |
| dc.format.extent | 1580308 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.source.uri (資料來源) | http://thesis.lib.nccu.edu.tw/record/#G1047510151 | en_US |
| dc.subject (關鍵詞) | 風險值 | zh_TW |
| dc.subject (關鍵詞) | 類神經網路 | zh_TW |
| dc.subject (關鍵詞) | GARCH模型 | zh_TW |
| dc.subject (關鍵詞) | 反向傳播演算法 | zh_TW |
| dc.subject (關鍵詞) | 共變異數法 | zh_TW |
| dc.subject (關鍵詞) | 歷史模擬法 | zh_TW |
| dc.subject (關鍵詞) | Value-at-Risk(VaR) | en_US |
| dc.subject (關鍵詞) | Neural Network | en_US |
| dc.subject (關鍵詞) | Variance-covariance method | en_US |
| dc.subject (關鍵詞) | Historical simulation method | en_US |
| dc.subject (關鍵詞) | GARCH(1,1) model | en_US |
| dc.subject (關鍵詞) | Backpropagation | en_US |
| dc.title (題名) | 以類神經網路建構風險值模型 | zh_TW |
| dc.title (題名) | Constructing a VaR model by Artificial Neural Network | en_US |
| dc.type (資料類型) | thesis | en_US |
| dc.relation.reference (參考文獻) | 英文文獻[1] Chen C.T. (2009), “Forecasting Value at Risk (VAR) in the futures market using Hybrid method of Neural Networks and GARCH model”, 2009 International Joint Conference on Computational Sciences and Optimization p17-21.[2] Moreno J.J.(2011),”Artificial neural networks applied to forecasting time series”, Psicothema 2011. Vol. 23 n^。2, p322-329.[3] Chen X.L(2009), “A Statistical Neural Network Approach for Value-at-Risk.”, 2009 International Joint Conference on Computational Sciences and Optimization p17-21.[4] Engle(1982), “Autoregressive Conditional Heteroscedasticity with Estimation of the Variance in U.K. Inflation,” Econometrica, Vol. 50, 1982, p987-1008..[5] Čorkalo Š (2011), “Comparison of a Value at Risk approaches on a stock portfolio”, Croatian Operational Research Review (CRORR) Vol. 2, 2011 p81-90.[6] Arnerić1 J.(2014), “GARCH based artificial neural networks in forecasting conditional variance of stock returns”, Croatian Operational Research Review p329-343.[7] Andjelic ́j G.(2010), “Application of VaR in emerging markets: A case of selected Central and Eastern European Countries”, African Journal of Business Management Vol. 4(17), p3666-3680.[8] Tsay R.S. (2002),” Analysis of Financial Time Series,” ISBN-13: 978- 0470414354[9] Kupiec(1995), “Techniques for Verifying the Accuracy of RiskMeasurement Models,” The Journal of Derivatives, winter, p73-84.[10] Rumelhart & Hinton(1986),“Learning representations by back-propagating errors”, Nature volume 323, p533–536.[11] Yan Liu(2005), “Value-at-Risk Model Combination Using Artificial Neural Networks”, Emory University. August 2005..中文文獻[1] 洪儒瑤、古永嘉、康健廷(2006)，ARMA-GARCH 風險值模型預測績效實證，Journal of China Institute of Technology Vov.34-2006.6。[2] 柯博倫、雷立芬(2010)， GARCH 估測風險值績效之探討，臺灣銀行季刊第六十二卷第四期(p234) 。[3] 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，第5卷第2期，19-42。[4] 蔡玉娟、林家妃、張修明(2010)，應用倒傳遞類神經網路及時間序列法建構股價報酬率預測模型-以台灣股市為例，國立屏東科技大學資訊管理系碩士班論文。[5] 林建甫(1996)，GARCH模型條件變異數結構變動的檢定，[6] 蘇榮斌、蔡孟祥(2008)，風險值之預測:以台灣、韓國、新加坡及馬來西亞等國家股票市場為例，Journal of China Institute of Technology Vol.39-2008.12[7] 許和鈞(2013)，利用風險值內部模型法提升證券商衍生性商品業務之風險控管效能，台灣證券交易所委託研究[8] 葉怡成(2003)，類神經網路模式應用與實作，臺北市：儒林。 | zh_TW |
| dc.identifier.doi (DOI) | 10.6814/THE.NCCU.MATH.007.2018.B01 | en_US |
