• Login
• Admin

Zhang, Ruone (2021) The Influence of Geo-Spatial Factors on Airbnb in New York City. [Dissertation (University of Nottingham only)]

PDF - Registered users only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Download (1MB)

Abstract

Airbnb has gained extensive popularity worldwide in only 13 years since its emergence. The company has grown at a surprising speed, and it has enabled millions of individual hosts to participate in the tourist accommodation sector by revolutionizing traditional peer-to-peer lodging with a new technology-driven online platform, unlike the hotels whose location is often analyzed by professional organizations and are restricted in commercial zones. The distribution of Airbnb is more random and dispersed since the hosts can be anywhere. The distribution of Airbnb listings in New York City shows apparent spatial heterogeneity. It is worth revealing the determinants behind the pattern.

This study aims to reveal the influence of geo-spatial factors on the popularity of Airbnb listings in New York City. A complete dataset of 36230 Airbnb listings in New York City was collected from the Inside Airbnb website, and from which 7724 popular listings were identified. Besides, geographical data of subway stations, bus stops, and commercial pedestrian zones were collected and calculated as additional features of each listing.

Two types of linear models were applied in this study. One is the General Linear Model that assumes global parameter estimates over the whole region. The other is the Geographically Weighted Regression which estimates parameters for each site differently depending on spatial correlations among neighboring regions. Model results prove that the Geographically Weighted Regression works better in goodness-of-fit and parameters’ accuracy and interpretability.

The model reveals three geo-spatial features that Airbnb guests highly value. Primarily, the fast accessibility to Midtown Manhattan and Lower Manhattan should be within 30 minutes of transit trip length. Moreover, it would be better if multiple subway lines near the listing for convenient subway access to other places in the city. Finally, the listings are expected to be close to commercial districts or overlays where tourists can easily find restaurants and shops. Based on these patterns of popular listings, two types of locations for future Airbnb listings are recommended at the end of the paper.

Item Type:	Dissertation (University of Nottingham only)
Depositing User:	ZHANG, RUONE
Date Deposited:	25 Apr 2023 10:10
Last Modified:	25 Apr 2023 10:10
URI:	https://eprints.nottingham.ac.uk/id/eprint/66483

Actions (Archive Staff Only)

Edit View