Large Sample Estimators of the Stochastic Discount Factor
提出利用大截面个股数据估计随机贴现因子的方法，通过短时间块结构和偏差修正实现一致性，并应用于美国个股数据检验常见资产定价因子的风险溢价。
Abstract We propose estimators of the stochastic discount factor using large cross-sections of individual stocks. We introduce a short time-block structure on a large N, T panel to exploit unbalanced panels of individual stock returns and suggest a novel bias correction to achieve the consistency of our estimators. Our estimators can accommodate pre-specified traded and nontraded factors, and latent factors. The estimators perform well in simulations. We apply our estimators to return data for U.S. individual stocks over a 50-year sample period and identify those factors in popular asset pricing models that command significant premia. A number of proposed nontraded factors have insignificant risk premia. Contrary to many studies, we find the market factor has a significant premium, as do profitability, value, and momentum factors.