This project aims to create an accurate predictive model for housing prices in Philadelphia County, Pennsylvania, by analyzing a wide range of variables that influence property values. The study considers intrinsic and extrinsic characteristics of housing units, as well as the spatial process of prices, to provide valuable insights for investors, government agencies, and homeowners in the local housing market. By employing Ordinary Least Squares (OLS) regression and incorporating a spatial lag variable, the model seeks to capture the complex dynamics influencing housing prices and deliver more accurate predictions. The results indicate that the predictive model performs moderately well, explaining a significant portion of the variance in sale prices. Factors such as property condition, amenities, neighborhood characteristics, and socioeconomic variables are found to have a significant impact on property values. The spatial analysis reveals the presence of spatial autocorrelation, with properties of similar sale prices tending to cluster together. The study discusses the model's strengths, limitations, and potential enhancements, providing a foundation for understanding the determinants of property values in Philadelphia County and guiding pricing strategies for stakeholders in the local housing market.
