Exploring the Spatial Hedonic Pricing Models for Determining Land Value in Wakiso District, Uganda.

Abstract

The existing traditional land valuation methods do not take into consideration spatial variations and influence of heterogeneous factors affecting valuation. This results in inflated and poor-quality land valuations that do not match the market price. One of the reasons for poor value estimation is use of methods that do not effectively analyze the massive amount of spatial data required to estimate the land value. Therefore, the purpose of this study was to explore the use of Geographic Information Systems (GIS) as a tool in determining Land value using the spatial Hedonic Pricing Models (HPMs). The study adopted GIS tools for multilinear regression modelling to investigate the relationship between land sales price and factors influencing land value. A total of 234 plots of land from five real estates in Wakiso district were used as the sample size, and attribute data for each plot was collected through ground survey done with the global positioning system (GPS) and document review. Ten explanatory variables were identified and utilized in the modeling exercise. Spatial regression models (SRMs) were used to link the ten inherent and neighborhood plot attributes to the plot market value at the time of sale. By using ordinary least squares (OLS) and SRMs various tests were conducted to identify whether there was multicollinearity, heteroscedasticity or spatial autocorrelation among variables. It was found that the spatial error model (SEM) stood out as the best land valuation model for the study area compared to the spatial lag model (SLM). Thus, the SEM was used to determine the Hedonic pricing model parameters for Land valuation in the study area and the model was tested on 16 plots sold in year 2022 in the same study area. The resultant GIS based Hedonic Land Valuation Model (HLVM) indicate that 79.7% land value can be determined using the following key significant parameters; i.e. plot size, plot’s distance to major roads, plot’s distance to major town and plot’s average slope. The SEM analysis also identified the plot size as the main positive influence of the land value while the other factors provides the least in influence. The results also proved that the GIS based HPM can effectively improve transparency in land valuation process since it can be analytically explained to landowners why a parcel was assigned a certain land value. GIS tools were employed to capture, manage, organize, analyze, visualize and present plots’ spatial information. GIS also played an important role in efficiently extracting spatial variables and lessen labor and time input. Therefore, Land valuers should apply GIS based HPMs in determining the significant and intrinsic factors in land valuation process to improve the quality of land value estimates.