Forecasting US Commercial Property Price Indexes Using Dynamic Factor Models

The general purpose of a dynamic factor model (DFM) is to summarize a large number of time series into a few common factors. In this paper we explore several DFMs on 80 granular, non-overlapping commercial property price indexes in the US, quarterly from 2001Q1 to 2017Q2. We examine the nature and the structure of the factors and the index forecasts that can be produced from the DFMs. We consider specifications of one to four common factors. As a major motivation for the use of DFMs is their ability to improve out-of-sample forecasting of systems of numerous related series, we apply the DFM estimated factors in an Autoregressive Distributed Lag (ARDL) model to forecast individual market index returns. We compare for four markets the forecasts to those from a benchmark univariate autoregression. The results show that the DFM & ARDL model predicts the crisis and subsequent recovery really well, whereas the benchmark model typically extrapolates the past price trend.