A Foray Exploration of Gentrification Through Modelling Approaches: A Case Study of Washington D.C.

Abstract

Within the urban area, various phenomena occur, but some phenomena change urban areas from positive or negative perspectives. Gentrification is one of such phenomenon, which arises impacts from both positive and negative perspectives. This work applies a difference-in-differences (DID) model and k-means clustering on real census data to explore different perspectives of the impacts of gentrification. The DID model is to evaluate the impact of the gentrification policy related to local resident welfare. The results of the DID model show that the increasing housing price is the positive impact of the gentrification policy. Following that, a conventional machine learning method: k-means clustering is utilized to identify the gentrified areas, which shows the neighborhood changes from 2010 to 2019.