Stock Price Jumps and Return Predictability

摘要

In a continuous-time framework, risk premium can only take the form of continuous drift and not the form of jumps. We use recently developed statistical techniques to identify jumps in stock prices, and examine the relation between jumps and cross-sectional stock return predictability associated with size, value, momentum, net stock issues, and liquidity effects. We find that stock price jumps can explain both the size and liquidity effects, and to a moderate extent the value premium. In contrast, the momentum and net stock issues effects are not driven by jumps. We further show that price jumps driving return predictability are not caused by regime shifts in risk. These findings provide an interesting perspective for evaluating risk-based and behavioral explanations of stock return predictability.