We propose a novel method for estimating the emotional returns of collectibles based on factor-mimicking portfolios. We leverage as much as 110 years of collector returns across 13 different asset classes and apply machine learning techniques to handle the challenges of non-synchronous trading. We use these estimates to look at how emotional returns affect equilibrium pricing. Emotional return estimates for twenty-four of our 30 collectible return series are positive, with an annual mean (median) of two.64% (2.53%). Despite various types of underestimation, these results provide evidence that assets with positive emotional returns have lower equilibrium financial returns.
