By Jacob Epstein
This paper explores the relationship between active mutual fund performance and market dispersion from January 1990 to December 2018. I find a significant positive relationship between dispersion and 4-factor alpha overall, providing some evidence of managerial skill. There are large differences in this relationship by decade and fund selectivity. The results suggest active mutual funds were able to take advantage of stock-picking opportunities during the 1990s and 2000s, particularly the most active subset of funds. However, I find a significant negative relationship between dispersion and alpha for funds in the 2010s, indicating this relationship has changed over time. I discuss several possible explanations for this reversal, which could present interesting avenues for further research.
Advisors: Professor Emma Rasiel | JEL Codes: G1, G12, G23
By Nalini Gupta
This paper seeks to test the hypothesis that developing countries or informationally inefficient countries should see higher returns for active mutual funds on average than passive funds and the trend should be reversed in developed nations or informationally efficient economies. This analysis is done using a cross section of eight countries, four developed and four developing. Using a fund universe of 20 active and 20 passive funds per country and controls such as volatility, market return, financial market development and Human Development Index among others, we see that there is no clear systematically dominant strategy between active and passive investment universally. While developing countries are associated with lower returns, we do not find a significant difference between active and passive based on development classification. A key finding is that an increase in liquidity, acting as proxy for informational efficiency, leads to a co-movement of active and passive returns in each country. The paper also lends itself to further analysis regarding confounding factor such as noise trading and movement of foreign capital which impact the effect of increased liquidity on mutual fund returns.
Advisors: Professor Connel Fullenkamp, Professor Kent Kimbrough | JEL Codes: G1, G11, G14
Forecasting Corporate Bankruptcy: Applying Feature Selection Techniques to the Pre- and Post-Global Financial Crisis Environments
By Parker Levi
I investigate the use of feature selection techniques to forecast corporate bankruptcy in the years before, during and after the global financial crisis. Feature selection is the process of selecting a subset of relevant features for use in model construction. While other empirical bankruptcy studies apply similar techniques, I focus specifically on the effect of the 2007-2009 global financial crisis. I conclude that the set of bankruptcy predictors shifts from accounting variables before the financial crisis to market variables during and after the financial crisis for one-year-ahead forecasts. These findings provide insight into the development of stricter lending standards in the financial markets that occurred as a result of the crisis. My analysis applies the Least Absolute Shrinkage and Selection Operator (LASSO) method as a variable selection technique and Principal Components Analysis (PCA) as a dimensionality reduction technique. In comparing each of these methods, I conclude that LASSO outperforms PCA in terms of prediction accuracy and offers more interpretable results.
View Data (Email for Access)
Advisors: Professor Andrew Patton, Professor Michelle Connolly | JEL Codes: G1, G01, G33
By Peter Noonan
This thesis analyzes factors that determine acquisition premiums paid by private equity firms in public to private leveraged buyouts. Building off of established literature that models the acquisition premiums paid in corporate mergers and acquisitions (M&A), this paper considers factors that influence a private equity firm’s willingness to pay (referred to as reservation price) and the bargaining power dynamic between a target company and acquirer in leveraged buyouts. Specifically, multivariable regression analysis is used to quantify the impact of a target company’s trading multiple, profitability, stock price as a percentage of its annual high, and number of competitors, a private equity firm’s deal approach and payment method, and the financial market’s 10-year US Treasury yield and high-yield interest rates at the time a transaction was announced. A sample of 320 public to private leveraged buyout transactions completed from 2000 to 2020 is constructed to perform this paper’s regression analysis. Using 2008 as an inflection point, this thesis then applies the same regression model to the subperiods from 2000–2008 and from 2009–2020 to examine how these drivers have changed as a result of industry trends—increased competition, low interest rates, and new value creation investment strategies—as well as the 2008 financial crisis and US presidential election—two crucial events that caused tremendous change in the financial system and intense scrutiny of the private equity industry. From the same original transaction screen, a second sample of 659 transactions is used to perform a difference of acquisition premium means t-test to analyze how the absolute magnitude of leverage buyout acquisition premiums have changed across these two subperiods. The second sample consists of more transactions due the t-tests less data-demanding nature as a result of its fewer variables. Results of this paper’s baseline model suggest that acquisition premiums are driven by a target company’s…
Advisors: Professor Ronald Leven, Professor Michelle Connolly | JEL Codes: G3, G11, G34
By Maria Suhail and Cipriano Echavarría
This thesis contributes to existing knowledge of private equity (PE) by analyzing the
impact of PE ownership post-IPO upon the long-term performance of companies. It considers whether companies perform better when PE funds maintain their ownership stakes post-IPO and whether this performance is also impacted by the degree of ownership that is maintained after IPO. This study uses stock performance (measured by cumulative excess stock returns) as a proxy for long-run company performance. The paper constructs and analyzes a sample of 487 companies that underwent an IPO between 2004 and 2012 to determine the implications of the maintenance and level of PE ownership by analyzing the performance of these companies for six years post-IPO. Results suggest that PE ownership post-IPO positively impacts long-term stock performance of companies. Duration and degree of PE ownership post-IPO are also important determinants of long-run performance likely due to the positive signal that continued PE ownership sends to outside investors about the quality of the company, the information asymmetry that exists between public and private markets and that PE firms are experienced managers that add value to companies.
Advisors: Professor David Robinson, Professor Michelle Connolly | JEL Codes: G11, G14, G24
By Tyler Fenton and Jarred Kotzin
The traditional efficient market hypothesis serves as the foundation of modern economic theory, governing the investigation of financial markets. While this premise assumes all investors are rational and all information is immediately incorporated into markets, this paper explores herding behavior – a central tenet of behavioral finance that explains the apparent inefficiencies of financial markets. Utilizing return data from the past 10 years from eight exchanges around the world, segmented into 10 industry classes as well as a broad market index, we compare levels of herd behavior using return dispersion proxies. We find significant evidence of herding in nearly all exchanges and all industries included in the study and the degree of this herd behavior varies across industries in different countries. Overall, we find support for the behavioral finance principle of herding and conclude that certain cultural or non cultural factors affect this activity differently in various countries and industries.
Advisors: Professor Connel Fullenkamp | JEL Codes: G4, G14, G15
By Matthew J. Farrell
Do hedge funds earn statistically significant premia on common factor trading strategies after trading costs are accounted for? Furthermore, what is the gap between what a hedge fund would earn and the paper portfolios that they hold? I answer this question by using the latest cutting-edge methodology to estimate trading costs for major financial market anomalies. This methodology uses the familiar asset-pricing Fama-MacBeth procedure to compare the on-paper compensation to factor exposures with those earned by hedge funds. I find that the typical hedge fund does not earn profits to value or momentum, and and low returns to size.
Advisor: Professor Brian Weller | JEL Codes: G12; G14; G23;
By Weiran Zeng
Prediction in economics is the focal point of debate for the future of economics, ever since economists were burdened with the failure to “predict” the 2008 Financial Crisis. This paper discusses positions held by philosophers and economic methodologists regarding what kinds of predictions there are and creates a taxonomy of prediction. Through evaluation of those positions, this paper presents different senses of prediction that can be expected of economics, and assess economists’ reflections according to those senses.
Advisor: Kevin Hoover | JEL Codes: B41, N1, G17
By Rafal Rokosz
The advent of blockchain technology has created a new asset class named cryptocurrencies that have experienced tremendous price appreciation leading to speculation that the asset class is experiencing an asset bubble. This paper examines the novelty and functionality of cryptocurrencies and potential factors that may lead to conclude the existence of an asset bubble. To empirically evaluate whether the asset class is experiencing an asset bubble the LPPL model is used. The LPPL model was able to successfully identify two of the four crashes within the data set signifying that cryptocurrencies are within an asset bubble.
Advisors: Ed Tiryakian and Grace Kim | JEL Codes: G12, Z00, C60
By Brian Perry-Carrera
With the recent growth in the investment of cryptocurrencies, such as bitcoin, it has become increasingly relevant to understand what drives price formation. Given that investment in bitcoin is greatly determined by speculation, this paper seeks to find the econometric relationship between public sentiment and the price of bitcoin. After scraping over 500,000 tweets related to bitcoin, sentiment analysis was performed for each tweet and then aggregated for each day between December 1st, 2017 and December 31st, 2017. This study found that both gold futures and market volatility are negatively related to the price of bitcoin, while sentiment demonstrates a positive relationship.
Advisor: Grace Kim | JEL Codes: G12, G41, Z00