Modeling Credit Risk in a Reduced Form Model
Date Issued
2010
Date
2010
Author(s)
Hsieh, Ya-Ting
Abstract
In the recent decades, literatures on credit risk measurement evolved dramatically. According to modeling techniques, they can be roughly grouped into two major categories, “structural models” and “reduced form models.” This thesis is based mainly on reduced form models and connects with structural models which include the advantages of the economic intuition to set up two models of evaluating credit risk, Jarrow04 Model and Jarrow09 Model. With the criteria of no-arbitrage pricing of bonds, pricing formulae of the two proposed reduced form models are derived for zero-coupon bonds. This research takes samples from zero-coupon bonds issued by the two major U.S. enterprises. Under the company’s information observed by market, evaluate the two zero-coupon bonds. The models can simulate many random variable paths and then the distributions of random variable. With variable distribution and the pricing formulae of zero-coupon bonds, we can obtain the bond prices in future periods. Furthermore, forecast errors are showed to compare the accuracy of valuing in the two models. We find that Jarrow04 Model has the best prediction ability for long-term zero-coupon bond during the financial crisis, and Jarrow09 Model predicts better for long-term zero-coupon bond not in financial crisis. In addition, for short-term bond, since it is close to risk-free bonds, there is no difference between the two models. Therefore, we suggest that using Jarrow09 Model to valuate zero-coupon bond is more convenient.
Subjects
Credit Risk
Reduced Form Model
Recovery Rate
Default Intensity
Zero-coupon Bond
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-99-R96221051-1.pdf
Size
23.53 KB
Format
Adobe PDF
Checksum
(MD5):2880e5c57dd545bde65af66a31a37440