債務違約機率之評估與CDS定價─台灣公司債市場之模擬

Abstract

CDS(信用違約交換，Credit Default Swap)是以各種債務的信用風險為標的之衍生性商品。此商品雖在歐美等國家行之有年，卻尚未於台灣被核准交易。但以CDS在全球之龐大的交易量而言，此商品未來在台灣被核准交易是可期待的。除了介紹CDS之運作與市場概況，本研究主要是結合CDS定價公式，與三種評估公司債違約機率的模型，計算以台灣公司債為參考實體之CDS合理定價。 本研究所使用的違約機率評估模型，包括了LS-LT、KMV兩種結構式模型，與普瓦松─指數分配之簡化模型。結構式模型是以分析公司之財務結構，而得到公司於未來特定時間內的違約機率。簡化模型是以債券每日市場交易價格，分析市場對該公司營運狀況之研判，並預測公司債未來違約之機率。 本研究設定台灣之公司債為CDS合約之參考實體，並且設定不同到期時間，以計算這些債務CDS之合理定價。最後，三種模型之計算結果，以及其間差異之分析將於文末呈現。

CDS (Credit Default Swap) is a derivative based on the credit risk of all kinds of debt obligation. Although this product has already traded in European countries and United States for a few years, it is not yet authorized to be traded in Taiwan. But for the globally large traded volume, it is expected that CDS can be authorized in Taiwan in the future. Besides introducing the operation and the market condition of CDS, this research is major in integrating the CDS pricing formula and three kinds of default probability estimation models and computes the rational pricing of CDS which is based on the corporate bonds in Taiwan. This research used two kinds of structural form model, LS-LT, KMV, and a reduced form model, Poisson and exponential distribution model. The structural form model is to analyze the financial structure of the corporate to get the probability of default in a certain period. The reduced form model is to observe the market price of the bond to analyze the market judgement to the enterprise, and then predict the probability of default. This research set the corporate bonds in Taiwan as the reference entity of CDS contracts and set different maturity to compute the CDS pricing. Finally the result of computation from three models and the analysis of difference will be presented in the end of this article.