This repository contains the code and data related to the article titled "A Comparative Study of Spatial and Non-Spatial Modelling in Price Prediction: A Case Study of Airbnb Price Prediction in Amsterdam." This paper provides a comprehensive analysis of the factors affecting Airbnb prices in Amsterdam and contrasts the predictive accuracy of spatial and non-spatial models.
Airbnb is a prominent example of how the sharing economy and peer-to-peer activities have acquired popularity in recent years. As Airbnb continues to expand, concerns have arisen regarding its effect on local economies and housing costs. For policymakers and the housing market, it is therefore essential to comprehend the factors that influence Airbnb pricing.
This study concentrates on predicting the price of Airbnb listings in Amsterdam using two models: the Hedonic Price Model (HPM) and the Globally Weighted Regression (GWR) model. The HPM is a global model that determines the transactional price based on intrinsic attributes, whereas the GWR model takes into consideration local variation and spatial dependence.
-
source: Contains the source code for data collection, preprocessing (implemented in Python), and model implementation (implemented in R). -
data: The dataset (both raw and pre-processed) can be found here. -
paper: Paper_Case Study of Airbnb Price predictions.pdf - This PDF file contains the full paper elaborating on the methods, results, and analysis and discussion of the study.