Daily House Price Indices: Construction, Modeling, and Longer‐run Predictions

Abstract

SUMMARYWe construct daily house price indices for 10 major US metropolitan areas. Our calculations are based on a comprehensive database of several million residential property transactions and a standard repeat-sales method that closely mimics the methodology of the popular monthly Case-Shiller house price indices. Our new daily house price indices exhibit dynamic features similar to those of other daily asset prices, with mild autocorrelation and strong conditional heteroskedasticity of the corresponding daily returns. A relatively simple multivariate time series model for the daily house price index returns, explicitly allowing for commonalities across cities and GARCH effects, produces forecasts of longer-run monthly house price changes that are superior to various alternative forecast procedures based on lower-frequency data.

A Re-Assessment of House Price Indices: Evidence From the Spanish Market

“…For example, Wu, Deng, and Liu (2014) argued that the simple average method fails to take account of housing quality adjustment, nevertheless these authors observed that in emerging housing markets such as China the initial price index was based on a weighted average method. Furthermore, such approaches have received applicability in mature markets, with the median value method used by the National Association of Realtors in the US (Bollerslev, Patton, & Wang, 2016).…”

“…The second method is the hedonic with prices corrected by quality through identifying attribute contribution to price, or shadow prices, and isolating the pure impact on prices. This is the main advantage of the hedonic approach but a fundamental challenge, again observed across the literature, is the requirement for extensive data sets (Clapman, Englund, Quigley, & Redfearn, 2006) given the infrequency of sale (Wood, 2005;Rappaport, 2007;Bollerslev et al, 2016). The functional form in a hedonic equation (Equation 1) is akin to regression, where P is the log of transaction prices, x it is the 'i' house attributes matrix, b i1 and b 0 are the parameters to be estimated.…”

“…Third, that houses which resale frequently tend to appreciate at higher rates. Fourth, that short holding periods may capture significant improvement to properties thereby violating assumptions on constant quality (Maguire et al, 2016;Bollerslev et al, 2016;Goh et al, 2012). Indeed, Goh et al (2012) identified that of five different models/variations they assessed, the repeat sales model was the least preferred.…”

Bank-Specific Shocks and House Price Growth in the U.S.

“…It is constructed using a repeat-transactions methodology, which is an increasingly common practice in housing research (Bollerslev et al, 2016). The FMHPI index is estimated with data including transactions on one-family and townhome properties serving as collateral on loans originated between January 1, 1975 and the end of the most recent index month.…”

Does Temporal Aggregation Explain the Persistence of the S&P/Case‐Shiller Indices? Evidence from a Longitudinal Specification

“…Recently, Bollerslev, Patton and Wang () have introduced a daily transaction‐based repeat sale index methodology that uses the Kalman filter (as opposed to a moving‐average rule) to smooth out noisy prices from home transaction data. Like Cotter and Roll, the researchers assume that the true index follows a random walk and conclude that “the dynamics dependencies in the new daily housing price series mimic those of other financial asset price.”…”