Pricing strategy on AirBnb based on the Spatial Influence of Touristic Attractions in New York
Resumo
This study evaluates variables supporting pricing strategies for hosts interested in competing on a sharing economy icon dominating markets in many cities, AirBnb. Data created by Airbnb worldwide offers a rich repository for insights on the real estate realm by crossing multiple variables, including some with geographical characteristics that can support the decision making process of millions of property owners looking for obtaining profit out of their properties. New York City (NYC) is one of the most visited cities in the world and data published by the Airbnb website contains variables that support deciding whether Airbnb can be an revenue stream option for property owners. The methodology consisted of exploratory quantitative analysis (spatial point pattern, density maps of listings, evaluation of block proximity to attractions in price using buffer analysis) followed by spatial dependence models: spatial auto-correlation and spatial regression models (SAR), using socio-economic-demographic variables and distances from NYC attractions to properties. The key results indicate important moderate spatial auto-correlation of rental prices (Moran´s I of 56.85%) and the consequent increase of explanation of rental prices in the models of 21.3 percentage points with the incorporation of variables of distance to NYC attactions and of the auto-regressive spatial term (R2 from 34.59% of OLS regression to 55.89% of the Spatial Autoregressive model).
Trilha Acadêmica: Big Data and Applied Retail Analytics