Exploring nonlinear and interaction effects of TOD on housing rents using XGBoost
Document Type
Journal Article
Publication Date
2025
Subject Area
place - asia, place - urban, land use - impacts, land use - planning, land use - transit oriented development, land use - urban density, policy - equity, policy - sustainable
Keywords
transit-oriented development (TOD), sustainable urban development
Abstract
Understanding the relationship between transit-oriented development (TOD) and housing rents is crucial for formulating effective TOD strategies and optimizing housing market management. These strategies contribute to a healthy housing market and sustainable urban development. Traditional regression models used in existing studies often fail to capture the nonlinear, and interaction effects of TOD on housing rents. This study addresses these limitations by applying the eXtreme Gradient Boosting (XGBoost) algorithm combined with Shapley Additive Explanations (SHAP) analysis to evaluate the effects of TOD on housing rents within Wuhan's Third Ring Road. Our approach not only identifies key TOD factors such as overall walkability, parking lot density, and commercial density but also uncovers significant nonlinear and threshold effects on housing rents. Moreover, we reveal the intricate interaction effects among key TOD variables, demonstrating how the local impact of one factor can be amplified or diminished by changes in another. This study provides novel insights into the complex mechanisms of TOD impacts on housing rents and offers actionable guidance for crafting targeted urban development strategies that promote urban equity and foster a sustainable housing market.
Rights
Permission to publish the abstract has been given by Elsevier, copyright remains with them.
Recommended Citation
Peng, C., Yang, S., Zhang, P., & Hu, S. (2025). Exploring nonlinear and interaction effects of TOD on housing rents using XGBoost. Cities, 158, 105728.

Comments
Cities Home Page:
http://www.sciencedirect.com/science/journal/02642751