Investor Behavior and Stock Returns

Theoretical and empirical studies usually assume that agents are all rational thinkers in making decisions. Experimental evidence is, however, that human judgments are not always rational, and people make systematic mental mistakes by using some intuitive simplifying rules and shortcuts, rather than strict logic to make choices. Judgment errors can lead to decisions that differ from those that would be made by rational agents. These shortcuts are also known as "behavioural heuristics".

This thesis contributes to the existing literature by exploring the influence of behavioural heuristics on the first four moments of the stock return distribution. It is comprised of three research papers. In the first two research papers, behavioural heuristics are accounted for through the subjective probabilities that an investor assigns to future stock returns. The third paper uses a different approach to recognize the investor's irrationality. It incorporates different behavioural heuristics through various behavioural factors that describe investors' biased investment decisions during extreme market conditions.

The first paper "Investor Behaviour and the Predictability of Stock Returns" (coauthored with A. Semenov) examines the implications of behavioural heuristics for the ability of past returns to predict future returns on individual stocks. We find empirical evidence that assuming investor's rationality, one may substantially understate (overstate) the predictability of the next period stock return when the next period volatility of returns is higher (lower) than the historical volatility.

The second paper "Behavioural Value-at-Risk" explores whether behavioural heuristics have the power to explain the volatility of stock returns. We find strong evidence that availability heuristic together with either optimism or overconfidence provides a more accurate forecast of the portfolio VaR compared with the conventional historical simulation and weighted historical simulation approaches.

Finally, the third paper "Higher-Order Return Moments under Irrational Behaviour" (coauthored with A. Semenov) explores whether investors' irrational behaviour through various behavioural factors influences the third (skewness) and fourth (kurtosis) moments of the stock return distribution. Empirical evidence demonstrates that the asymmetric distribution of stock returns due to the existence of negative skewness and excess kurtosis can be explained by the investors' overreaction to price signal, herding intensity, and momentum return.