Essays in Predictive Empirical Finance

Open Access
- Author:
- Henrichsen, Aaron Peter
- Graduate Program:
- Business Administration
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- March 01, 2017
- Committee Members:
- Jingzhi Huang, Dissertation Advisor/Co-Advisor
Jingzhi Huang, Committee Chair/Co-Chair
Tim Simin, Committee Member
David Haushalter, Committee Member
John Liechty, Outside Member - Keywords:
- Exchange Traded Funds
Market Prediction
Commodities
Diversification
Governance - Abstract:
- This dissertation contains three essays in empirical finance which each use predictive statistics to analyze characteristics of firms, market conditions, or funds to anticipate future events. In the first, I study the information content of flows of assets into leveraged exchange traded funds. Using daily flow and return data, I find that flows to leveraged ETFs predict subsequent benchmark index returns. These results remain significant to a lesser extent at the weekly level, but disappear for monthly returns. Separating ETFs into commodity, debt, domestic equity, foreign equity, and real estate sectors, I find that out-of-sample predictability is limited to foreign equity and junk bonds. A self-funding strategy based on out-of-sample predictability earns over 5\% per year, but the majority of this is negated by transaction costs. This suggests that although flows do not predict most market sectors, flows can anticipate movements in markets where there are higher transaction costs or less liquidity. In the second study, I investigate the dynamics of equity and commodity market correlation. Unlike previous economic downturns where equity returns dropped but commodity futures returns largely stayed positive, following the financial crisis equity and commodity futures returns showed much higher than normal levels of co-movement. Using mean-variance spanning tests, I find that though diversification from a portfolio of solely equities into a portfolio of equities and commodities generally improves an investor's efficient frontier, it did not yield any significant improvement from 2008 to 2010 and little to no improvement in 2007 and 2011. Frontiers post crisis starting in 2012 are again benefited from diversification into commodities. Using quality of predictive power for several factors, I find that the levels of global supply and demand, and to a lesser extent commodity relevant hedge fund assets under management (AUM), and index investment aid in predicting future commodity-equity correlation leading into the financial crisis as well as during the subsequent recovery. Finally, in the third study I examine the associations of various types of director expertise on a distressed firm’s ability to avoid bankruptcy. Since the advent of the Sarbanes Oxley Act of 2002 there have been significant increases in the percentage of board members with accounting, investment banking and legal expertise. In a study of US firms during the period between 1996 and 2011, I find that the percentages of bankers and lawyers on boards are associated with significant changes in the likelihood that a firm will enter bankruptcy in the future. No similar effect is observed for board member accounting expertise. These results support the idea that bankers may aid distressed firms in renegotiating their debt, enabling them to avoid bankruptcy. Conversely, a board with more legal expertise may be instrumental in steering a firm through bankruptcy.