This thesis examines the behavior of professional and individual traders and its relationship with trading profits, return characteristics, and asset prices. I first examine this through the impact of hedge fund managers' public announcements on stock prices and fund returns. I then focus on individual trading patterns and the relationship between trading, return and stock characteristics and trader profitability.
In "Synchronized Arbitrage and the Value of Public Announcements", I test the idea that arbitrageurs use public announcements as a synchronizing signal. I find that firms publically identified by hedge fund managers as being overvalued underperform their respective benchmarks by 324 to 376 basis points per month, during the 24 months subsequent to the public announcement. In contrast, firms identified by managers as being undervalued do not overperform their benchmarks. I find evidence of coordination by arbitrageurs through an increase in post-event short selling and changes in funds' derivative positions. Finally, I find that the long-short return disparity cannot be resolved by common explanations, such as varying manager skill, short sales constraints, analyst downgrades or slow information diffusion, but is rather derived from managers' idiosyncratic ability to predict future accounting deficiencies and negative earnings news.
In "Investor Overconfidence, Turnover, Volatility and the Disposition Effect: A Study Based on Price Target Updates", I analyze the interaction between investor overconfidence, share turnover, return volatility and the disposition effect. I create a measure for investor overconfidence using posted price target updates retrieved from Yahoo!Finance online message boards. I find that posters are slow to adjust price targets to changes in underlying stock prices. These updating patterns appear to be suboptimal when contrasted with realized return and sell-side analyst benchmarks. I further find that the overconfidence exhibited by posters is strongly associated with share turnover, return volatility, and the magnitude of the disposition effect. The results in this paper add to our understanding of the link between behavioral biases and investor and stock-level anomalies in financial markets.
In "It's Not the Trader, It's the Trade: A Cross-Sectional Analysis of Individual Traders' Activity", I examine the effect of the holding period on stock selection preferences by individual investors. I find that individual investors' preferences regarding volatility are highly correlated with their expected holding period. Short term trades are associated with high beta, high standard deviation, high M/B stocks, while long term trades are associated with lower beta, lower standard deviation, lower M/B stocks. I also show that per-trade abnormal returns are positively correlated with the expected holding period. Short term trades provide, on average, significant negative abnormal returns while long term trades produce positive abnormal returns. Finally, I show that risk preferences among individual investors are cross-sectionally consistent, i.e. that preference changes cannot be explained by a specific group of traders.