Mexico city's airbnb listing price analysis using regression Academic Article in Scopus uri icon

abstract

  • © Proceedings of the 13th IADIS International Conference ICT, Society and Human Beings 2020, ICT 2020 and Proceedings of the 6th IADIS International Conference Connected Smart Cities 2020, CSC 2020 and Proceedings of the 17th IADIS International Conference Web Based Communities and Social Media 2020, WBC 2020 - Part of the 14th Multi Conference on Computer Science and Information Systems, MCCSIS 2020. All rights reserved.The AirBnb platform provides users with the option of renting their vacant spaces as tourist accommodations and competing with traditional accommodation enterprises. However, since AirBnb gives the user the freedom to establish the price of their listing, a challenge is placed on the owner to determine the most appropriate number. This study analyses the information from listings in Mexico City to determine how the listing attributes can be used as predictors for a new listing's price. The study uses statistical methods and machine learning techniques to analyze the information scraped from AirBnb's website, which is publicly available at the Inside AirBnb webpage. In this work, an experiment was made to compare a quantile regression, logistic regression, and a generalized additive model to find the most suitable technique for predicting an AirBnb listing's price. The models were compared based on the residual standard error, R squared, and AIC. The results show that the generalized additive model provides the best fit for the dataset explaining 60% of the variance.

publication date

  • January 1, 2020