Spatial Variance Modelling for Price Values of Residential Lands in Riyadh

Abstract

The increasing population and urban growth in Riyadh city during the last decades have been accompanied by an increasing demand for residential and a rise in their prices to meet the growing demographic and economic needs (Royal Commission for the City of Riyadh, 2023). This study aims to estimate the price of land and its distribution in Riyadh by comparing three different spatial regression models: ordinary least squares (OLS), Geographically Weighted Regression (GWR), and Multi-scale Geographically Weighted Regression (MGWR). These models were employed in this study to identify the pattern of the relationship between the price of land and the number and density of the population. Sixty-one land parcels offered for sale were randomly selected as case study. This study illustrates that the MGWR spatial regression model is distinctly better than the OLS and GWR models. Furthermore, this study also reveals that the highest land prices are concentrated along the highway networks closest to downtown.

Keywords: land prices, spatial regression models, OLS, GWR, MGWR, Riyadh, Saudi Arabia