Comparison of reduced form and structural credit risk models

The institution for my research visit is the New York University`s Leonard N. Stern School of Business. NYU Stern has taken its place among the most distinguished business schools worldwide where more than 200 faculty members undertake research on the very top level. An important part of NYU Stern is its world-renowned finance department. It has been ranked number one world-wide in research productivity by a number of top finance journals. Stern`s finance faculty is highly rated in terms of research output, and its members serve on the editorial boards of all of the major finance journals. My supervisor and faculty mentor is Marti G. Subrahmanyam, Charles E. Merrill Professor of Economics and Finance. I have chosen NYU Stern because it provides the optimal environment for my research project. The aim of my research project is to contribute to the finance literature, especially to the credit risk literature. The management of credit risk is the key challenge of financial institutions nowadays. Traditionally, credit risk positions are built up by granting loans or trading in bond markets. In the last ten years markets for credit risk derivatives started to evolve providing the opportunity to actively manage credit risk exposures. Since then these markets have shown tremendous growth rates which surpassed even the boldest predictions. The availability of adequate pricing models is therefore of key importance for financial institutions. In my research project I will analyze the result from calibrating state-of-the-art reduced form and structural credit risk models to a broad set of market data. The fast growth of credit derivatives markets provides an important new set of price information. I have high quality data for bond and credit default swaps provided by Markit Group Limited at hand and I will add stock market and balance sheet data. This broad data set enables more powerful parameter estimation and allows for the calibration of more complex models. Academic studies so far have either not used credit derivatives data for model calibration or have only tested very particular specifications. A comprehensive study comparing different models which uses especially credit derivatives data as input is still missing. My research project shall provide important insights for the understanding of credit risk models considering credit derivatives data as the major data source and shall discuss possible model extensions.