博碩士論文 105221029 完整後設資料紀錄

DC 欄位 語言
DC.contributor數學系zh_TW
DC.creator林晏瑜zh_TW
DC.creatorYen-Yu Linen_US
dc.date.accessioned2018-7-25T07:39:07Z
dc.date.available2018-7-25T07:39:07Z
dc.date.issued2018
dc.identifier.urihttp://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=105221029
dc.contributor.department數學系zh_TW
DC.description國立中央大學zh_TW
DC.descriptionNational Central Universityen_US
dc.description.abstract在保險業中我們經常需要做隨機模型情境的模擬,藉由過去的數據來模擬分析未來發生的所有可能性,並進行現金流量分析,然而因龐大的模擬次數,現金流量測試大量的運算時間成為了一項很大的挑戰。因此為了解決大量運行時間的問題,本研究利用不同抽樣方法在不減其模型有效性下減少樣本情境數。本文首先以幾何布朗運動模擬出一百萬組股價報酬的情境,再建立三種不同抽樣方法分別進行抽樣,檢視這三種抽樣方法是否能夠有效減少模型運行次數,並比較三種抽樣方式的優劣,且同時檢驗抽樣後隨機模型的有效性。最後本研究將抽樣結果實際應用在隨機現金流量測試中,利用條件尾端期望值作為標準,對保單責任準備金的適足性進行測試,來達到運用抽樣方法減少情境次數同時大幅減少現金流量測試責任準備金適足性運行時間的目的。zh_TW
dc.description.abstractIn the insurance industry, simulations of stochastic model scenarios are frequently used to simulate and analyze future possibilities with data in the past and conducted in cash flow testing. However, due to the huge number of simulation times, long time of computing cash flow testing becomes a big challenge. Therefore, to solve the problem of long computing time, this research reduces the number of scenario with different sampling methods but under the premises that no efficiency of models would be affected. In this paper, we use geometric Brownian motion to simulate scenarios for ten thousand groups of stock price return, and then we build three different sampling methods to check the efficiency of stochastic modeling from scenario sampling. Finally, we apply the sampling result to the stochastic cash flow testing. In addition, we use conditional tail expectation as a standard to test the adequacy of policy reserve and thus to reduce scenario times, and at the same time, to reduce cash flow testing time by using scenario sampling.en_US
DC.subject抽樣方法zh_TW
DC.subject情境抽樣zh_TW
DC.subject隨機現金流量測試zh_TW
DC.subjectsampling methoden_US
DC.subjectscenario samplingen_US
DC.subjectstochastic cash flow testingen_US
DC.title隨機模型情境抽樣有效性之應用zh_TW
dc.language.isozh-TWzh-TW
DC.titleThe Application of Efficient Stochastic Modeling from Scenario Samplingen_US
DC.type博碩士論文zh_TW
DC.typethesisen_US
DC.publisherNational Central Universityen_US

若有論文相關問題,請聯絡國立中央大學圖書館推廣服務組 TEL:(03)422-7151轉57407,或E-mail聯絡  - 隱私權政策聲明