Areas of Expertise (5)
Machine Learning in Finance
Professor Clark's research focuses on financial intermediation and risk management with an emphasis on machine learning. He currently teaches several quantitative finance courses including Financial Computation and Simulation and Advanced AI/ML for Finance.
Additionally, Clark has spent a decade working as a Senior Financial Economist with the US Office of the Comptroller of the Currency.
Clark's work has been published in the Journal of Banking and Finance, Journal of Financial Stability, and Operations Research Letters.
Rensselaer Polytechnic Institute: Ph.D., Finance
Clarkson University: MBA
Clarkson University: M.E., Mechanical Engineering
Rensselaer Polytechnic Institute: B.S., Biomedical Engineering
Media Appearances (2)
How Will Banking Be Affected by the Presidential Election Result?
U.S. News & World Report print
The U.S. president has a lot of power to shape the policies that determine how banks do business. Many of these policies affect consumers through their bank accounts, credit cards, loans, and other financial products and services. So the way banks operate could be affected by which man wins November's election, President Donald Trump or former Vice President Joe Biden. "The president has broad power to set banking regulations both directly and indirectly," says Brian Clark, assistant professor in the Lally School of Management at Rensselaer Polytechnic Institute.
Coronavirus may make cashless more common. Who does that leave out?
Albany Times Union print
"Retail is going away, and naturally people are shopping online for everything," said Brian Clark, an assistant professor at Rensselaer Polytechnic Institute's Lally School of Management. "But even once you go in-person, I think we’re already moving toward cashless anyway. A lot of it’s here, and I think the pandemic will probably accelerate things."
Risk and risk management in the credit card industryJournal of Banking & Finance
Florentin Butaru, Qingqing Chen, Brian Clark, Sanmay Das, Andrew Lo, and Akhtar Siddique,
Using account-level credit card data from six major commercial banks from January 2009 to December 2013, we apply machine-learning techniques to combined consumer tradeline, credit bureau, and macroeconomic variables to predict delinquency. In addition to providing accurate measures of loss probabilities and credit risk, our models can also be used to analyze and compare risk management practices and the drivers of delinquency across banks. We find substantial heterogeneity in risk factors, sensitivities, and predictability of delinquency across banks, implying that no single model applies to all six institutions. We measure the efficacy of a bank's risk management process by the percentage of delinquent accounts that a bank manages effectively, and find that efficacy also varies widely across institutions. These results suggest the need for a more customized approached to the supervision and regulation of financial institutions, in which capital ratios, loss reserves, and other parameters are specified individually for each institution according to its credit risk model exposures and forecasts.
A Machine Learning Efficient FrontierOperations Research Letters
Brian J. Clark, Zachary Feinstein, and Majeed Simaan
We propose a simple approach to bridge between portfolio theory and machine learning. The outcome is an out-of-sample machine learning efficient frontier based on two assets, high risk and low risk. By rotating between the two assets, we show that the proposed frontier dominates the mean-variance efficient frontier out-of-sample. Our results, therefore, shed important light on the appeal of machine learning into portfolio selection under estimation risk.
Bank loan renegotiation and credit default swapsJournal of Banking & Finance
Brian Clark, James Donato, Bill Francis, and Thomas Shohfi
Using Roberts (2015) loan-level data from 2000 to 2011, we find that the inception of CDS trading on reference firms’ debt is associated with a decreased number and lower probability of amendments, restatements, and rollovers to existing lenders of bank loans. Reference firms are also less likely to terminate loans prematurely or refinance with different lenders after the inception of CDS trading and tend to exhibit longer loan maturities. Our evidence is consistent with the empty creditor problem arising from CDS trading and the resulting decrease in the negotiation power of borrowers. Our research contributes to understanding how financial innovations alter bank-lending relationships.
Consumer Defaults and Social CapitalJournal of Financial Stability
Brian Clark, Iftekhar Hasan, Helen Lai, Feng Li, and Akhtar Siddique
Using account level data from a credit bureau, we study the role that social capital plays in consumer default decisions. We find that borrowers in communities with greater social capital are significantly less likely to default on loans, even after adjusting for different levels of income and other characteristics such as credit scores. The results are strongest for potentially strategic defaults on mortgages; a one standard deviation increase in social capital reduces such defaults by 12.4 %. These results can be generalized to any mortgage default. Our results also indicate that the effect of social capital is most prominent among more creditworthy borrowers, suggesting that when given a choice, the social cost of defaulting is an important factor affecting default decisions. We find a similar impact of social capital on consumer defaults in other datasets with more detailed information on borrowers as well. Our results are robust to modeling and methodology choices, as well as controlling for other drivers of default such as wealth, income and amenities from homeownership. Our results suggest that increasing social capital via measures to build community cohesion such as promotion of owner-occupied home ownership may be one avenue to deter consumer default.