Essays in Financial Economics

Open Access
Author:
Huang, Zhijian
Graduate Program:
Business Administration
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
April 25, 2008
Committee Members:
  • Jingzhi Huang, Committee Chair
  • Jeremy Ko, Committee Chair
  • Jean Helwege, Committee Member
  • Runze Li, Committee Member
Keywords:
  • Asset pricing
  • data snooping
  • market efficiency
  • out-of-sample test
  • behavioral finance
Abstract:
Previous studies show mixed results about the out-of-sample performance of various asset-pricing anomalies. To reduce data-snooping bias, this paper simulates a real-time trader who chooses among all asset-pricing anomalies published prior to that time using only non-forward-looking filters. I find that a trader can outperform the market by recursively picking the best past performer among published anomalies. The excess return tends to be highest when the trader looks at past performances between two years and five years and when the trader considers more anomalies. For published anomalies, their excess returns over benchmark as well as relative ranks among contemporaneous anomalies do not decrease over time, indicating a relatively stable performance once being published. Relying only on the then-available anomaly literature and historical data, the overall result shows a possible way to beat the market in real time. In the second essay, I study the managerial decision and its impact on mutual fund performance. ``Meet or beat" describes manager's manipulation of the accounting earnings to meet analyst forecasts, which causes the final earnings distribution to be negatively-skewed with a dent to the left. I investigate if mutual fund managers also demonstrate such a risk-seeking behavior and the potential impact on mutual fund performance. I develop a two-period decision model in which a manager makes a decision to take or not to take a risky project when she faces a target level of performance in time 1. The manager's payoff depends on a linear function of the time 2 (final) value of the firm, and on a step function on whether the firm meets the target or not. As a result, those managers with firm values slightly lower than expected tend to take the risky project, and if the firm value is already higher or far below the expectation, the manager does not take the project. The terminal value of a firm conforms with the empirical evidence of meeting or beating analyst forecasts. I empirically test the model on actively managed mutual funds. I find funds in the 3rd and 4th worst performance deciles take higher risks and on average outperform the median decile in the following year, indicating a risk seeking behavior for slightly under-performing managers. I also compare the excess return distributions for growth funds and index funds, but do not find negative skewness supporting the model. The third essay is an experimental study about people's time-inconsistent risk preferences. There is a substantial literature which studies time-inconsistent temporal preferences. We conducted an experiment to explore time-inconsistency in the other dimension of investment preferences, i.e., risk preferences. We had subjects play a multi-period betting game where they planned their betting decisions in advance and then played the game dynamically later to see if these decisions matched their plan. We found that subjects took more risk than planned in their initial bet and after a loss where this increase in risk-taking is associated with an increase in breakeven mental accounting. Our findings shed light on the conditions under which emotions exacerbate mental accounting and other behavioral biases.