Determinants of Airbnb prices in European cities: A spatial econometrics approach (Supplementary Material)

<p>This repository contains supplementary materials for the article:</p> <p><strong>Determinants of Airbnb prices in European cities: &nbsp;A spatial econometrics approach</strong></p> <p><strong>(</strong>DOI<strong>:&nbsp;</strong><a href="https://doi.org/10.1016/j.tourman.2021.104319">https://doi.org/10.1016/j.tourman.2021.104319</a>)</p> <p>The materials include the used datasets and Python&nbsp;scripts for spatial regression models.</p> <p><strong>Datasets</strong></p> <p>For each city two files are provided: data for weekday&nbsp;and weekend offers</p> <p>The columns are as following:</p> <ul> <li>realSum: the full price of accommodation for&nbsp;two people&nbsp;and two nights in EUR</li> <li>room_type: the type of the accommodation&nbsp;</li> <li>room_shared: dummy variable for shared rooms</li> <li>room_private: dummy variable for private rooms</li> <li>person_capacity: the maximum number of guests&nbsp;</li> <li>host_is_superhost: dummy variable for superhost status</li> <li>multi: dummy variable if the listing belongs to hosts with 2-4 offers</li> <li>biz: dummy variable&nbsp;if the listing belongs to hosts with more than 4 offers</li> <li>cleanliness_rating: cleanliness rating</li> <li>guest_satisfaction_overall: overall rating of the listing</li> <li>bedrooms: number of bedrooms (0 for studios)</li> <li>dist: distance from city centre in km</li> <li>metro_dist: distance from nearest metro station in km</li> <li>attr_index: attraction index of the listing location</li> <li>attr_index_norm: normalised attraction index (0-100)</li> <li>rest_index: restaurant&nbsp;index of the listing location</li> <li>attr_index_norm: normalised restaurant&nbsp;index (0-100)</li> <li>lng: longitude of the listing location</li> <li>lat: latitude of the listing location</li> </ul> <p><strong>Programming&nbsp;Scripts</strong></p> <p>In this repository you will find a script for spatial regressions in Python using PySAL (models_robust.py).</p> <p>The codes cover the following regression models:</p> <ul> <li>OLS</li> <li>SLX (lagged_x)</li> <li>SAR (lagged_y)</li> <li>SDM (lagged_x_y)</li> <li>SEM (lagged_e)</li> <li>SDEM (lagged_e_x)</li> </ul> <p>Main parameters:</p> <ul> <li>cities - list of cities from the dataset to be included in the analysis</li> <li>Robust=False: calculate the OLS, SLX, SAR and SDM regressions with W (weight matrix) based on 10 closest neighbours</li> <li>Robust=True: calculate all regression models with different specifications of W</li> <li>direct_indirect=True: calculate the direct and indirect effects (based on Golgher, A. B., &amp; Voss, P. R. (2016). How to Interpret the Coefficients of Spatial Models: Spillovers, Direct and Indirect Effects. Spatial Demography (Vol. 4). https://doi.org/10.1007/s40980-015-0016-y)</li> </ul> <p>Key functions:</p> <ul> <li>create_weights - defines the W specification</li> <li>write_stats - calculates&#39;s Moran&#39;s I and Geary&#39;s C</li> <li>direct - calculates the direct effect of the variable</li> <li>indirect - calculates the indirect effect</li> <li>coord - sets the coordinate refence system (CRS) appropriate to the analysed city</li> <li>total_results calculates the regressions</li> <li>the coordinates are projected from GPS (epsg:4326) to the local CRS (km_lat, km_lon)</li> <li>all regressions are saved as formatted txt table</li> <li>the results can be also saved as csv table</li> </ul> <p>&nbsp;</p> <p>&nbsp;</p> <p>&nbsp;</p> <p>&nbsp;</p> <p>&nbsp;</p>