Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/24861| Title: | 應用Copula方法於厚尾財務資料及風險值之計算 The Application of Copula Methods to Heavy-Tailed Financial Data and the Computation of Value at Risk The Application of Copula Methods to Heavy-Tailed Financial Data and the Computation of Value at Risk The Application of Copula Methods to Heavy-Tailed Financial Data and the Computation of Value at Risk |
| Authors: | Wen-Hua Hsieh 謝文華 |
| Advisor: | 葉小蓁(Hsiaw-Chan Yeh) |
| Co-Advisor: | 陳宜良(I-Liang Chern) |
| Keyword: | 連接函數,一般化柏拉圖分配,風險值,厚尾, Copula,Generalized Pareto Distribution,Value at Risk,Heavy Tail, |
| Publication Year : | 2007 |
| Degree: | 碩士 |
| Abstract: | 在這篇論文中，我們應用連接函數與極端值理論配置AMD及XEROX兩間公司股票報酬的聯合分配。我們比較了一般化柏拉圖分配以及常態分配作為邊際分配的影響，並且計算95%, 99%, 99.5%, 99.9%, 99.95% 及99.99% 信賴水準的風險值。由連接函數及極端值理論所得到的風險值的估計值，不論在樣本內或樣本外的測試結果都很好。並且我們也發現風險值的估計值並不隨著柏拉圖分配門檻的選擇而有很大的變化。另一方面，常態分配作為邊際分配所得到的信賴水準99%以上的風險值則低估了實際的風險。 |
| URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/24861 |
| Fulltext Rights: | 未授權 |
| Appears in Collections: | 數學系 |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| ntu-96-1.pdf Restricted Access | 1.36 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.