Backcasting, Nowcasting, and Forecasting Residential Repeat-Sales Returns: Forecast Combination meets Mixed Frequency

Abstract

The Case-Shiller is the reference repeat-sales index for the U.S. residential real estate market, yet it is released with a two-month delay. We find that incorporating recent information from 71 financial and macro predictors improves backcasts, nowcasts, and short-term out-of-sample forecasts of the index returns. Combining individual forecasts delivers large improvements in forecast accuracy at all horizons. Additional gains are obtained with mixed-data sampling methods that exploit the daily frequency of financial variables, reducing the out-of-sample mean squared forecast error by as much as 11% compared to a simple autoregressive benchmark. The forecast improvements are largest during economic turmoil and throughout the 2020 COVID-19 pandemic period.

Keywords: Real estate, Case-Shiller, MIDAS, Forecasting, Big Data

JEL Classification: C22, C53, R30