Three essays on investments and empirical asset pricing

No Thumbnail Available

Meeting name

Sponsors

Date

Journal Title

Format

Thesis

Subject

Research Projects

Organizational Units

Journal Issue

Abstract

This dissertation consists of three essays on investments and empirical asset pricing. The first essay in Chapter 1--Portfolio Manager Ownership and Low-Risk Anomalies--examines the impact of portfolio manager ownership (i.e., skin in the game) in the mutual fund industry on the relation between risk (e.g., beta, idiosyncratic volatility or distress risk) and abnormal return in the stock market. Across a comprehensive set of strategies that buy high-risk stocks and sell low-risk stocks (i.e., low-risk anomalies), I show that negative alphas concentrate only among stocks with low ownership intensity. The findings are consistent with the conjecture that agency-issue-induced incentives entail excessive risk taking that distorts the riskreturn relation. In the second essay in Chapter 2--Flow Hedging and Mutual Fund Performance-- I study the cross-sectional differences in the flow-hedging behavior of U.S. active mutual funds and their implication for fund performance. I find that nearly half of the funds do not hedge against systematic flow risk, and I explain this result with a rational model in which informed managers receive more precise private signals about systematic flows. In the data, I confirm the model's prediction that funds having higher exposure to flow risk generate better risk-adjusted performance. In the last essay in Chapter 3--Out-of-Sample Performance of Factor Return Predictors--I investigate the ability of a comprehensive set of variables in predicting equity factor returns. I apply a variety of shrinkage methods on the predictor variables to forecast almost a hundred equity factors and document robust evidence of out-of-sample predictability and investment outperformance for factor timing strategies.

Table of Contents

DOI

PubMed ID

Degree

Ph. D.

Thesis Department

Rights

License