Testing Pricing Errors of Models with Latent Factors and Firm Characteristics as Covariances
扩展了统计提取潜在因子的方法，将具有预测能力的公司特征作为条件协方差（贝塔），并规定定价误差与公司特征的仿射变换正交，从而显著区别于现有文献，并强烈拒绝了零定价误差假设。
This paper extends the methodology of statistically extracting latent factors in settings with return-predictive firm characteristics as conditional covariances (betas) between returns and factors. The main feature is that the pricing errors (alphas) are specified to be orthogonal to the affine-transformed firm characteristics as the betas with one component of pricing errors lying outside the space spanned by the firm characteristics. The specification is shown to make substantial differences with the extant literature as the zero pricing error hypothesis is strongly rejected for various models with commonly used firm characteristics. This paper was accepted by Agostino Capponi, finance. Supplemental Material: Data are available at https://doi.org/10.1287/mnsc.2023.4768 .