Predicting the equity risk premium using the smooth cross-sectional tail risk: The importance of correlation
提出一种新的月度横截面尾部风险指标SCSTR，通过模拟和实证发现其能更好捕捉月度尾部风险，并在1964-2018年间显著预测股权风险溢价，表现优于历史风险溢价等常用预测因子。
I provide a new monthly cross-sectional measure of stock market tail risk, SCSTR, defined as the average of the daily cross-sectional tail risk, rather than the tail risk of the pooled daily returns within a month. Through simulations, I find that SCSTR better captures monthly tail risk rather than merely the tail risk on specific days within a month. In an extended period from 1964 until 2018, this difference is important in generating strong in- and out-of-sample predictability and performs better than the historical risk premium and other commonly-used predictors for short- and long-term horizons.