Break Risk
提出一种新方法，利用横截面信息快速检测影响大量股票的断点，从而更准确地预测股票收益，并基于现值模型解释断点的经济来源。
Abstract We develop a new approach to modeling and predicting stock returns in the presence of breaks that simultaneously affect a large cross-section of stocks. Exploiting information in the cross-section enables us to detect breaks in return prediction models with little delay and to generate out-of-sample return forecasts that are significantly more accurate than those from existing approaches. To identify the economic sources of breaks, we explore the asset pricing restrictions implied by a present value model which links breaks in return predictability to breaks in the cash flow growth and discount rate processes.