The rental housing market in Barcelona: a nonparametric analysis

Abstract

The quality of geolocation services and the amount of publicly available data on the internet have increased substantially in the last decade. This paper studies the Barcelonese rental real estate market by exploiting both geolocation services and user-generated public data. This allows for the creation of a dataset which is later analysed using kernel smoothing methods. The key ndings are that (i) rent prices are approximately log-normally distributed, that (ii) rent prices are lower in areas further away from the city centre, and that (iii) the previous point has to be complemented with a geospatial analysis that accounts for spatial heterogeneity.