Modelling the Airbnb listings' price in Lisbon using local spatial regressions

Open Access

Abstract

Dissertation presented as partial requirement for obtaining the Master's degree in Statistics and Information Management, with a specialization in Information Analysis and Management ; Sharing economy market, such as Uber and Airbnb, have been growing rapidly in the last few years, providing extra income to agents from the supply side, and low costs to those in demand side. Although its adoption provided benefits for stakeholders and to the global economy of the areas in which they are inserted, several authors and politicians have been referencing the negative externalities brought with it, such as an increase in rents and real estate prices and a decrease in hotels' revenue. However, most of the externalities pointed out, were not based on any empirical analysis. The aim of this study is to analyze Airbnb market within Lisbon municipality, focusing mainly the modelling spatial variation of Airbnb listings' price. For this purpose, it was employed an ordinary least square (OLS) model and a geographical weighted regression (GWR) model to identify the main factors affecting the Airbnb listings' price. The results showed that the GWR model performs better than the OLS model, and it allows assessing the local impact of the explanatory variables on the Airbnb listings' price. In conclusion, it was found that the price of the two types of Airbnb listings (entire home/apartments and private/shared rooms) are not affected by the same factors and that statistically significant differences varied across parishes within Lisbon municipality. Perhaps, there is a need to test if it is plausible to apply a unique regulatory policy considering Airbnb and Lisbon market as an aggregated concept or by Airbnb listing type and Lisbon parishes.