High-frequency predictability of housing market movements of the United States : the role of economic sentiment

Abstract

We analyze the ability of a newspaper-based economic sentiment index of the United States to predict housing market movements using daily data from 2nd August, 2007 to 19th June, 2020. For this purpose, we use a nonparametric causality-in-quantiles test, which allows us to test for predictability over the entire conditional distribution of not only housing returns, but also volatility, by controlling for misspecification due to nonlinearity and structural breaks. Our results show that economic sentiment does predict housing returns (unlike the conditional mean-based Granger causality test) and volatility, barring the extreme upper ends of the respective conditional distributions.