Date of Award


Degree Type


Degree Name

Doctor of Philosophy in Business Administration


Business Administration

First Advisor

Georges Tsafack


The purpose of my dissertation is to develop a new methodology to detect and identify international financial contagion, utilize machine learning to predict which countries are most susceptible to financial contagion while identifying the factors that facilitate it, and to analyze the post-crisis government response.

In the first manuscript, I identify a new methodology to detect which countries suffered contagion due to an external financial crisis. Most economist agree that contagion is a change in the dependence structure during bad times in international financial markets, however measuring and testing this change remains a challenging issue. Correlation is often used to assess contagion but suffers a volatility bias. Forbes and Rigobon (2002) propose a correction to this bias based on a strong assumption of constant beta. We find that this assumption is not supported by data. I then suggest the rank correlation as an alternative measure which not only is robust to volatility bias but is free of assumption. Using that measure, I test the contagion for a large number of financial crises, and unlike Forbes and Rigobon, I find contagion in most cases.

In the second manuscript, I utilize an empirically defined dataset detecting international financial contagion, this paper utilizes machine learning via a Random Forest algorithm to provide a proof-of-concept for predicting the spread of contagion. Focusing on capital flows, direction of trade, regional, and other idiosyncratic features, this paper finds that a forecast accuracy of over 70% is possible. Ultimately, the largest barrier to improving the forecast accuracy is a matter of reliable, consistent international data.

In the third manuscript, I analyze the long-term effects of the Home Affordable Refinance Program (HARP), specifically the likelihood of delinquency for a borrower that refinanced into the HARP program versus a borrower that did not, despite the fact they were qualified to have done so. As a response to the financial crisis, the Home Affordable Refinance Program (HARP) was passed in March, 2009. Since then, over 3.4 million borrowers have refinanced through the HARP program. We investigate the impact of this program on the mortgage performance. We find that the program leads to a decrease of delinquency rates. However, while the loans refinanced through the program perform relatively well in the short-run following their refinancing, the positive impact tends to revert over time.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.