22/08/2026
🏙️ What can short-term rental data tell us about the forces shaping urban housing markets?
Day 6 of our BIP continued with Professor Adriana Davidescu and PhD student Diana Agafitei from ASE Bucharest, who led the workshop:
“Urban Rent Dynamics: Geospatial & Predictive Analysis with Python”
Using real Inside Airbnb data from European cities, students followed a complete analytical pipeline, from cleaning and questioning raw, messy data to developing a defensible rental-price model.
They created interactive geospatial maps, investigated host concentration and professionalization, and compared Linear Regression, Ridge Regression and Random Forest models. Using SHAP, they also explored how to interpret model predictions and understand the influence of location, room type and host behaviour on rental prices.
The workshop demonstrated that predictive accuracy is only part of the story. Transparent and interpretable models can also generate meaningful economic insights into urban rent dynamics, housing markets and regional development.
Thank you to our two professors for an engaging, accessible and strongly applied workshop connecting Python, geospatial analysis and economic research!
cc: