******************************************************************************

                                    Seminar


       Department of Systems Engineering and Engineering Management,

                    The Chinese University of Hong Kong

Title:
On the Systematic Bias of Testing Structural Models of Credit Risk

Speaker:
Professor Hoi-Ying WONG
Statistics Department
The Chinese University of Hong Kong

Date : March 3, 2006 (Friday)

Time : 4:30 p.m. - 5:30 p.m.

Venue : Room 513, William M.W. Mong Engineering Building

(Engineering Building Complex Phase 2), CUHK

Abstract:
Empirical studies so far suggested that structural models generated a
significant pricing error for corporate bonds, implying that these models
failed to explain credit risk. In fact, barrier-independent models were shown
to underestimate corporate bond yields substantially while barrier-dependent
models overestimate the yield on average but contain a sizeable amount of
underestimation. We argue that these findings are bias since the proxies used
to replace model parameters conflict with the model feature. Therefore, no
matter how satisfactory of the theoretical feature of a model, its empirical
use has been limited by the chosen implementation method. We use options'
properties to prove the biasness contained in the proxies and propose that
parameters are better estimated with statistical techniques. Two interesting
results are presented. First, we derive implications of using the proxy for
firm value in estimating default barriers. The implications deduce that the
barrier option framework of Brockman and Turtle (2003, Journal of Financial
Economics) on estimating default barrier is significantly biased upward.
Empirical evidence is provided to further illustrate the effect. Second, we
spell out the systematic bias embedded in the empirical frameworks in testing
corporate bond pricing in the past. Besides giving a mathematical
justification, maximum likelihood estimation is applied to conduct an
empirical study on the Merton, LS and CDG model. Our empirical results show
that structural models perform much better with statistical estimation and do
not consistently underestimate bond yields.

Bio:
Hoi Ying Wong obtained his PhD degree in Mathematics from Hon Kong University
of Science and Technology. His research interests include Stochastic Models,
Estimation Methods, and Mathematical and Statistical Finance with special
attention to Derivative Pricing, Interest Rate Modeling and Financial Risk
Management. He is an assistant professor in the Department of Statistics of
the Chinese University of Hong Kong. He was the director and one of the
founders of the Master of Science in Risk Management of Chinese University of
Hong Kong.

Note : Cookies and drinks will be available at 4:15 pm.

 _______________________________________________________________________________