This paper presents a hedonic housing price model for the city of Glasgow in Scotland. The major innovation of the research is the use of hierarchical clustering techniques to identify property submarkets defined by a combination of property types, locations and socioeconomic characteristics of inhabitants. Separate hedonic price functions are estimated for each submarket and these functions are shown to differ significantly across submarkets. Further, the paper illustrates the use of a generalised moments estimator proposed by Kelejian and Prucha that accounts for spatial autocorrelation in property prices. Spatial autocorrelation is shown to be an important consideration with this data. The principal motivation of the research is to provide an indication of the impact of road traffic noise on the market price of property. In all but one of the submarkets exposure to road traffic noise is shown to have a significant negative impact on property prices.