Geographic Clustering Analysis of the Impact of COVID-19 Pandemic on Housing Prices in the United States
The study employs Moran’s I statistic and Local Indicators of Spatial Association (LISA) to analyze spatial patterns and dependencies in housing prices across U.S. counties. It aims to understand how the pandemic affected housing prices in different regions in the U.S. and offers insights into pandemic-induced shifts in the real estate market.
Processed housing data covering 3,005 U.S. counties using Excel for data preparation. Utilized Python libraries such as GeoPandas and Matplotlib to create Choropleth maps showing the average annual housing prices and price growth rates in each county, highlighting spatial clustering phenomena in U.S. housing prices.
Employed Python's PySAL and Splot libraries for global and local Moran's I analysis of housing prices, and created LISA cluster maps. This revealed significant spatial autocorrelation in U.S. housing prices and the specific spatial differences in the impact of the pandemic on housing prices across regions.
Integrated data analysis results with relevant literature review to interpret spatial patterns in housing data, and proposed underlying reasons for changes in housing prices during the pandemic from both demand and supply perspectives.