Disentangling spatial and structural drivers of housing prices through Bayesian networks: A case study of Madrid, Barcelona, and Valencia.
An object of class bn.fit
. Refer to the documentation of bnlearn
for details.
A discrete Bayesian network modeling both spatial and structural drivers of house prices in the city of Madrid. The Bayesian network is learned as in the referenced paper. The vertices are:
Air conditioning available (Yes, No);
Age of the building (New Development, Modern, Mid-Age, Historic);
Constructed area in square meters (Small, Medium, Large, Luxury);
Number of bathrooms binned (Few, Moderate, Many));
Distance to city centre (Very Near, Near, Medium, Far);
Property condition (New Construction, Second Hand Renovation, Second Hand Good Condition);
Number of dwellings in the building (Low Medium, High, Very High-Density);
Doorman service present (Yes, No);
Floor level in the building (Lower, Mid, Upper, Top);
Garden present (Yes, No);
Distance to green space (Very Near, Near, Medium, Far);
Maximum building height (Low-Rise, Mid-Rise, High-Rise, Skyscraper);
Lift presente (Yes, No);
Distance to supermaket (Very Near, Near, Medium, Far);
Distance to metro station (Very Near, Near, Medium, Far);
Neighborhood frequency - quartile rank (Most Common, Frequent, Less Frequent, Rare);
Swimming pool present (Yes, No);
Asking price per square meter - binned (Very Low, Low, Medium Low, Medium High, High, Luxury);
Parking space available (Yes, No);
Cadastral building quality index (Low Value, Moderate Value, High Value, Very High Value);
Number of rooms - binned (Few, Moderate, Many);
Distance to primary avenue (Very Near, Near, Medium, Far);
Distance to secondary avenue (Very Near, Near, Medium, Far);
Storage room available (Yes, No);
Terrace present (Yes, No);
Property type (Studio, Duplex, Penthouse, Standard);
Built-in wardrobes (Yes, No);
Murga, A. G., & Leonelli, M. (2025). Disentangling Spatial and Structural Drivers of Housing Prices through Bayesian Networks: A Case Study of Madrid, Barcelona, and Valencia. arXiv preprint arXiv:2506.09539.