Exploring, modelling and predicting spatiotemporal variations in house prices

Abstract

Hedonic price modelling has long been a powerful tool to estimate house prices in the real estate market. Increasingly, traditional global hedonic price models that largely ignore spatial effects are being superseded by models that deal with spatial dependency and spatial heterogeneity. In addition, many novel methods integrating spatial economics, statistics and geographical information science (GIScience) have been developed recently to incorporate temporal effects into hedonic house price modelling. Here, a local spatial modelling technique, geographically weighted regression (GWR), which accounts for spatial heterogeneity in housing utility functions is applied to a 19-year set of house price data in London (1980London ( -1998 in order to explore spatiotemporal variations in the determinants of house prices. Further, based on the local parameter estimates derived from GWR, a new method integrating GWR and time series (TS) forecasting techniques, GWR-TS, is proposed to predict future local parameters and thus future house prices. The results obtained from GWR demonstrate variations in local parameter estimates over both space and time. The forecasted future values of local estimates as well as house prices indicate that the proposed GWR-TS method is a useful addition to hedonic price modelling.

Modeling dynamics of price changes in housing market in Kazan

“…While the authors developed models (3,4) in 2012, with a forecasting period up to the end of 2014, accuracy of forecasting can be predicted by comparing with actual data for the period from August 2012 till December 2014 (Fig. 1).…”

“…Therefore, it is of no surprise that there are a great number of authors concentrated on the analyses of the housing market problems. The various indexes and counting models are used in prognosis of the future housing prices, and the development of the certain housing policy [3,4] There are the scientists who are interested in the housing price formation process directly. They research the assets, and the correlation of the rent and housing prices [5].…”

Towards Increasing Residential Market Transparency: Mapping Local Housing Prices and Dynamics

“…The spatial structure of location features may also be included in the geographically weighted regression (GWR) model [36,37]. Due to price volatility over time, Fotheringham et al [38] postulate the use of GWR not only for spatial analyses but also for spatial and temporal analyses. These analyses allow us to present not only the current spatial distribution of prices and values but also local trends and short-term forecasts.…”

Mapping fine‐scale urban housing prices by fusing remotely sensed imagery and social media data

“…In addition, Bitter et al () also consider the spatial variance of housing prices and conduct geographically weighted regression on house sales data to predict housing prices and obtain higher accuracy than the hedonic housing price model; Kuntz and Helbich () utilize statistical data combined with the kriging interpolation method with consideration of houses' structural and neighborhood characteristics to map real estate prices. Considering the existing spatiotemporal heterogeneity and autocorrelation of housing prices, Wu, Deng et al () and Fotheringham, Crespo, and Yao () adopted detailed housing statistical data to simulate housing price distribution, which fully takes into account that different influence factors should be considered in different areas. As we know, these data sources have several main problems: first is their high cost, especially high labor cost and long update cycle; moreover, these data are discrete and have difficulty demonstrating the fine spatial patterns of city housing price distributions (Chen et al, ; Wu et al, ).…”