Disparities in Affecting Factors of Housing Price: A Machine Learning Approach to the Effects of Housing Status, Public Transit, and Density Factors on Single-Family Housing Price

Profound insights have been gained into which characteristics determine housing prices. These characteristics reflect two different aspects: those which are correlated with the dwelling itself and those which are correlated with the location and the surrounding area. However, few studies precisely looked at the disparities and heterogeneity in these effects across neighborhoods with varied conditions. Also, there lacks studies focusing on the moderate-density cases where housing markets have drawn concerns recently. This study aims to fill this research gap by analyzing these disparities across neighborhoods with different economic and racial/ethnic conditions. Through machine learning approaches, we compare the disparities in the impacts of housing status, public transit services, and surrounding environment factors under seven conditions. Results indicate that the heterogeneity in economic conditions could be more significant than racial/ethnical conditions. Through comparison analysis, we call policymakers to need to adopt differentiated perspectives on housing price analysis, and future studies should consider the disparities in the impacts across neighborhoods.