Machines and Masterpieces: Predicting Prices in the Art Auction Market
32 Pages Posted: 20 Mar 2019
Date Written: March 5, 2019
We study the accuracy and usefulness of automated (i.e., machine-generated) valuations for illiquid and heterogeneous assets. We assemble a database of 1.1 million paintings that were auctioned between 2008 and 2015. We use a popular machine-learning technique—neural networks—to develop a price prediction algorithm based on both non-visual and visual artwork characteristics. Our out-of-sample valuations predict auction prices dramatically better than valuations based on a standard hedonic pricing model. Moreover, they help explaining price levels and sale probabilities even after conditioning on auctioneers’ pre-sale estimates. Machine learning is particularly helpful for assets that are associated with higher levels of ex-ante price uncertainty. Finally, we show that it can help overcome experts’ systematic biases in expectations formation.
Keywords: asset valuation, auctions, experts, big data, machine learning, computer vision, art
JEL Classification: C50, D44, G12, Z11
Suggested Citation: Suggested Citation