Corporate default prediction & Credit Risk chain modelling

Yangzhengxuan (Monica) Wang is on a mission. While most credit risk studies ignore systemic risk, Wang believes this approach is unrealistic and potentially dangerous: “The sub-prime credit crisis perfectly illustrates the importance of this issue,”she says. “Traditional credit risk analysis led to inaccurate credit ratings and the consequent negative impact on mortgage lenders, major banks, and ultimately, the entire global financial market!”
Wang posits that credit risk can be more accurately measured by considering the system as a whole. To create a quantitative model for the credit risk chain between all the counterparties in the system, she has identified an optimal set of corporate default drivers, using them to evaluate 12 corporate default models and to improve the prediction ability of two random default measurement models: mixed logit and frailty.
Her strong background in financial mathematics has played a key role in Wang’s research. “My mathematical modelling skills have significantly helped my project, which covers 12 quantitative models with five model evaluation methods,” she says. “My maths background is the rock of my work: it has deepened my understanding of previous research and has driven me to go further in default risk prediction.”
Wang’s work is beneficial to a broad range of actors in the financial sector and society at large. “The findings from this project may help regulators and policymakers to better assess financial stability, creditors and bankers to maximise profits, auditors to more accurately evaluate firm health, investors to reduce portfolio losses, and shareholders to optimise returns,” she concludes. “Default risk is a risk for society, so developing better ways of measuring it benefits us all.”
My research focuses on investigating the factors which lead to default risk. Those factors include not just firm based information, but also macroeconomic indicators. The second part intends to explain and capture the default cluster and correlated default risk with advanced random models. Unobserved factors are tested and the default cluster is divided into industry, macroeconomic and calendar scope. The last part concentrates on evaluate prediction ability for varying models and suggest the best prediction model for further application.

Seeing the forest, not the trees

Biography
I am currently a PhD student of finance in School of Business and Economics, the University of Exeter. I graduated first from the Applied Mathematics Department of Shanghai University of Finance and Economic in 2007. During my university period, I was awarded several prizes of both national and international mathematical modeling competitions, which aroused my keen interest of doing research and modeling. In the summer of 2008, I was honored the Master degree of Science in financial mathematics with distinction. I did an internship in Galaxy Security Corporation of China 2007.

Research work
My research is about the credit risk management and it is supervised by Richard D F. Harris, a leading expert in risk management and financial econometrics. The objective of this project is to find a better way to measure and evaluate credit risk for the whole financial system. In particular, a quantitative model will be constructed for the credit risk chain between all of the counterparties in the system.

AXA funding
I first heard of AXA fellowship from my supervisor, Richard D F. Harris. I felt so exiting and lucky when I received the fantastic news that I was awarded the fellowship. It is confidence to say that this fellowship is good for my CV. As an international student, AXA shares my burden of life, which could defiantly help me to do my research better. Further, the kind and generous fellowship plan of AXA will influence my rest life that I would like to try my best to help others in the future.

To add or modify information on this page, please contact us at the following address: community.research@axa.com