Real-Time Forecasts of Auction Prices
30 Pages Posted: 26 Feb 2020
Date Written: October 5, 2014
Abstract
This paper investigates the informational content of aggregate prices in the fine arts auction market. A Mixed Data Sampling (MIDAS) modeling approach is proposed to forecast year-end art prices, using higher frequency variables related to the stock and bond markets and to art market sentiment. It takes about six months for art prices to incorporate information contained in the price of Sotheby's stock. Market sentiment variables such as trading volume have better explanatory power in the short-term. These findings suggest that art market participants react with a delay to information contained in stock market returns, so that information diffuses only slowly into art prices.
Keywords: Information and market efficiency; MIDAS; Information diffusion; Art auctions; Forecasting; Market sentiment.
JEL Classification: C58, G12, G14, Z11.
Suggested Citation: Suggested Citation