Securitization of Multi-Country Longevity Risk: a Dynamic Copula Approach

[[abstract]]本研究主要貢獻是將狀態轉換關聯模型運用在多國死亡率隨機模型之關聯結構。為捕抓 死亡率模型的厚尾高峰現象，本研究亦運用Wang et al. (2011)的模型，在假設死亡率指 數符從跳躍擴散分配或是廣義雙曲分配下，運用非常態之Lee Carter模型來配適各個國家 的死亡率資料。在死亡率關聯結構的應用方面，運用EIB/ BNP長壽債券之架構，本研究 設計一個報酬型態與存活指數有關的長壽債券，並在Wang風險中立轉換下，提供此長 壽債券的合理價值。此外，本研究未來亦將提供此長壽債券的風險值與條件風險值。運 用長年期之多國死亡率資料，本研究將探討多國死亡率之非對稱死亡關聯結構。最後， 運用數值分析，本研究將進一步探討非常態特性與死亡關聯結構對長壽債券之訂價與風 險控管之影響。[[abstract]]The main contribution of this project is to employ regime switching copula models with a tranquil regime and a co-jump regime to introduce multi-country mortality dependence in stochastic mortality modeling. In addition, following Wang et al. (2011), this paper also fits the original Lee–Carter model with the jump diffusion (JD) and generalized hyperbolic (GH) innovations to each mortality data. As an application, following the innovative EIB/ BNP longevity bond idea, I will design a longevity bond linked to a cohort survivor index and provide the fair values of longevity bonds using Wang transform. I will also provide the value at risk (VaR) and conditional tail expectation (CTE) of the longevity bonds numerically. Finally, using the calibrated parameters of multi-country long-term mortality data, I will provide the numerical analyses to demonstrate the asymmetric dependence phenomenon in multi-country mortality data and to assess the impacts of the non-normal property and asymmetric mortality dependence on the pricing and risk management of the longevity bonds.[[note]]NSC101-2410-H327-02