利率連動債券評價應用之分析

Abstract

本論文是分析利率連動債券評價工具的應用，其主要分析的商品包括：異幣別利率交換商品(Quanto Interest Rate Swap)、區間應計型利率商品(Range Accrual Note)、目標利率達成償還型商品( Target Redemption Note)、固定到期利率交換型商品(Constant Maturity Swap)、以及異幣別固定到期利率交換型商品 (Quanto Constant Maturity Swap)。並針對上述商品，介紹其商品特性，適用之利率模型，Excel 輸入(出)之畫面(用VBA程式語言撰寫)，和最終的評價結果。以利金融業之理財專員，於面對客戶購買連動債券時之評價需求；或是金融交易部之交易員，交易該商品時可以試算其投資報酬率；進而可以活絡利率連動債券之市場。 最後的研究結論歸納如下： 1.利率連動債券，種類繁多，千奇百怪，不容易暸解，更不易評價。不論理財專員、個人或法人投資者、及金融業者，於交易該類商品時，皆應準備簡易之評價工具，而該工具應由獨立的財務工程或數理統計之專業機構來協助開發及驗證。 2.在開發新型商品的評價工具時，財務專家和資訊專才，必須通力合作，並共享對方領域之知識。在設計評價工具的過程中，必須考慮如何使理財專員或一般投資者，也覺得非常容易操作。由應避免太多的參數及太多的估計值需要使用者輸入，那評價工具的方便性及正確性也會更提高。 3.電腦運算能力，是決定吾人可以使用最適模型來評價的成功因素。電腦主要運算機制(CPU)、主記憶體、程式語言和資料庫，都會影響運算速度。唯有強大的電腦運算能力，才能提供吾人用最好的模型，計算出最佳的結果；效能較佳或等級較高的筆記型電腦，即可滿足運算需求。 4.一般而言，BGM模型加上適當的運用蒙地卡羅模擬，可以計算出較佳的評價結果。在少數狀況，單用BGM模型，會產生太多的中間運算過程，可能無法在可接受的運算時間內，得到正確的評價結果；此外，BGM模型須要輸入較多之歷史資料及參數，應於使用者試算前，即予輸入，方可事半功倍。 最後，建議主管機構，協助建立超然的獨立機構，協助金融業界及投資者，發展簡單易用的評價工具，並可接受委託評價新型的衍生性商品。有三點好處：首先，減少個別投資者，因投資於衍生性商品，產生的糾紛；其次，協助金融機構，評估其持有部位之市場風險，可適時、適當的調整其風險資產之部位；最終，可活絡金融市場，尤其是金融再仲介 ( Financial Re-Intermediation )之業務。

This study is to analyze the application of pricing model on six interest rate derivatives which include Quanto Interest Rate Swap, Range Accrual Note, Quanto Range Accrual Note, Target Redemption Note, Constant Maturity Swap, and Quanto Constant Maturity Swap. For these interest rate derivatives, this study briefs the product term sheet attributes, interest rate model of pricing, excel input(output) format (under VBA coded), and the final output of pricing model. The empirical results are: 1.The interest rate derivatives types are too many to understand their attributes and pricing. For the individual or financial institution investors, they need to call help from the independent financial engineering consultant or company to solve the pricing issues. 2.To develop a new type of interest rate derivatives price tool, the financial experts and the information technology geniors need to work together and share the domain know how. They also need to consider the easy use on end user side. If they can avoid too many parameters, and too man estimations input, the accuracy of the output will be more certain. 3.The computing power is the critical successful factor which we calculate the price by using the suitable interest rate model. The CPU capability, main memory, computer language, and data base, all count on the performance. Only powerful computing power can provide us to use the right model and come out the right solution. 4.In general, BGM model and Monte Carol Simulation can get more accurate output in pricing. However, BGM model sometimes generates too many intermediate calculation process. We may not get the reasonable price within limited computing time frame. Finally, we suggest the authorization to establish an institution to help investors in price the in new interest rate derivatives. There are three advantages. First, it reduces the misunderstanding between the investors and issuers. Second, it helps the financial institutions evaluate their market risk. Finally, it activeates the financial re-intermediation market.