Artificial Intelligence, Data Science & Machine Learning
Ozde Oztekin
Dr. Ozde Oztekin commenced her academic career at the University of Kansas and currently holds a position as a full professor and an Eminent Scholar at the Florida International University. Dr. Oztekin’s research primarily centers around corporate finance, international finance, and banking. A specific area of interest for her has been understanding the decision-making processes of corporate firms and financial institutions. Recently, Dr. Oztekin has delved into the application of AI in finance. This is highlighted by her work on the role of machine learning models in comprehending capital structure dynamics and her research analyzing the influence of COVID-19 policy responses on local economic and health outcomes. She has been a prolific contributor to various reputable journals and has displayed a strong commitment to quality education, evident from her diverse teaching portfolio that spans across multiple levels and modalities. Beyond her teaching and research roles, Dr. Oztekin is an active participant in the academic community, refereeing for several journals, undertaking editorial responsibilities, and engaging in conference activities.
Publications and Book Chapters:
1. Can machines learn capital structure dynamics?
This project investigates the potential of machine learning models in predicting leverage more accurately than linear models. The study showcases that machine learning models, especially random forests, significantly boost prediction accuracy. These models were found to identify a broader set of leverage determinants, highlighting factors like market-to-book, industry median leverage, cash and equivalents, and more. The research underscored machine learning’s capability to provide a more precise target estimation and emphasize the primary drivers of leverage adjustments.
Links: https://www.sciencedirect.com/science/article/abs/pii/S0929119921001954
2. The impact of COVID-19 and its Policy Responses on Local Economy and Health Conditions
In collaboration with Ali Gungoraydinoglu and Ilke Oztekin, this research utilizes machine learning techniques to delve into the consequences of U.S. states’ lockdown measures meant to curb the spread of COVID-19. The study aims to unveil the marginal health benefits juxtaposed with the economic ramifications associated with social distancing. Through our analytical models, it was observed that while lockdowns alleviated disease severity, they also precipitated significant economic downturns. The decline in health conditions further impacted the labor supply, financial stability, and overall economic productivity. Interestingly, while the economic fallout from lockdowns surpassed the economic repercussions directly from the disease, health conditions presented a more reliable predictor for economic contractions.
Links: https://www.mdpi.com/1911-8074/14/6/233
3. “Coronavirus Pandemic and its Economic and Human Capital Costs in US States”
This book chapter expands upon a previous study, exploring the balance between health outcomes and economic impacts in US states during the COVID-19 pandemic. Machine learning assessment indicated that disease severity is a stronger predictor of real local economic contraction than mandatory social distancing policy measures. The research underlines how state policy measures and health conditions influenced unemployment, business bankruptcy, and overall economic activity.