Empirical Analysis of House Price Determinants in Lagos Using Hedonic Price Modelling

This study constructs an empirical hedonic price model to identify and quantify the determinants of residential property prices in Lagos State, South West Nigeria, contributing to a significant gap in the academic real estate literature on African metropolitan housing markets. While hedonic pricing methodology is well established in Anglo-American real estate research, its systematic application to Lagos's heterogeneous property market presents unique methodological challenges related to data quality, market segmentation, and the quantification of qualitative neighbourhood attributes. The study assembles a transaction dataset of 480 residential property sales across twenty Lagos neighbourhoods from estate surveyor firms and property developer records covering 2020 to 2023. Structural, locational, and neighbourhood quality variables are constructed for each observation, including floor area, building age, number of bedrooms, distance to central business district, road accessibility score, neighbourhood security rating, utility availability index, school proximity, and flood risk classification. Ordinary least squares and spatial lag regression models are estimated to account for spatial autocorrelation in price residuals. Results show that floor area, security rating, road accessibility, and utility availability are the four most significant positive price determinants across the full sample. Distance to CBD exhibits a non-linear negative effect, with marginal distance penalties higher in middle-income than high-income neighbourhoods. The spatial lag model demonstrates significant positive spatial autocorrelation in price residuals, indicating neighbourhood price spillovers. The study provides the most comprehensive hedonic price evidence for Lagos to date and offers methodological guidance for future Nigerian property market research. Keywords: hedonic pricing, house price determinants, Lagos, residential property, spatial autocorrelation.