Misspecification-Robust Inference in Linear Asset-Pricing Models with Irrelevant Risk Factors
发现当模型包含与测试资产收益不相关的风险因子时，传统推断方法会高估其定价能力；提出的模型选择程序能有效剔除不改善定价能力的因子，并应用于多个流行模型，发现只有市场和账面市值比因子被定价。
This paper shows that in misspecified models with risk factors that are uncorrelated with the test asset returns, the conventional inference methods tend to erroneously conclude, with high probability, that these factors are priced. Our proposed model selection procedure, which is robust to identification failure and potential model misspecification, restores the standard inference and proves to be effective in eliminating factors that do not improve the model's pricing ability. Applying our methodology to several popular asset-pricing models suggests that only the market and book-to-market factors appear to be priced, while the statistical evidence on the pricing ability of many macroeconomic factors is rather weak.