## 研究報告

 標題 delta-geometric random walk跳躍模型下信用風險之估算與實證分析 102 1 2013/08/01 delta-geometric random walk跳躍模型下信用風險之估算與實證分析 The Estimation and Empirical Study of Credit Risk in delta-geometric random walk Model with Jump Process 王仁和 淡江大學財務金融學系 計畫編號：NSC102-2410-H032-026 研究期間：201308~201407 研究經費：581,000 行政院國家科學委員會 本研究考慮在信用風險的delta-geometric random walk模型下，引進跳躍過程，在一般均衡下給出歐式選擇權的評價公式，並用來估算違約破產機率，並探討跳躍過程參數與違約破產機率關連性與結構，進而改進歐式選擇權的評價公式及違約破產機率。 實證方面，從 Datastream, Option Metrics 及台灣經濟新報資料庫蒐集一些國內外股票上市公司的資料，為delta-geometric random walk跳躍模型來補抓金融風暴下風險結構，並估計其參數。利用市場股價及股票選擇權來估算違約破產機率，並用CAP與ROC曲線和實務常用的KMV模型、KRM模型做比較，提供信用風險管理者做為決策參考的依據。;In this research plan, we consider using general equilibrium to derive the European option pricing formula and evaluate the default probability of Credit Risk under the model of delta-geometric random walk with jump process. We discuss the relation and structure between probability of default and parameters of jump process, then improve the more precise option pricing formula and probability of default. Empirically, we can collect some prices on domestic and foreign stocks from Datastream, Option Metrics and TEJ databases. Use model of delta-geometric random walk with jump process to capture the structure of probability of default and estimate the unknown parameters of model. Then, use both the stock prices and option prices to estimate probability of default of the proposed model. Finally, plot the CAP and ROC curves to compare with popular KMV, KRM model. 中文