Finite-Length Spatiotemporal Modelling for Housing Price Network Spillovers.

Area of Science:

• Urban Economics
• Spatial Econometrics
• Network Analysis

Background:

• Understanding housing price contagion across cities is crucial for economic stability.
• Existing models often fail to capture the integrated spatiotemporal dynamics of housing price spillover.
• A novel approach is needed to analyze intercity housing price connections comprehensively.

Purpose of the Study:

• To develop an integrated model for mapping directed spillover pathways in urban housing prices.
• To analyze the spatiotemporal integration characteristics of housing price contagion across Chinese cities.
• To identify dominant spillover hubs and evolving network structures.

Main Methods:

• Embedding finite-length sequence correlation analysis (Correlation-Dependent Balanced Estimation of Diffusion Transfer Entropy, CBEDTE) into the gravity model (GM) to create the CBEDTE-GM integrated model.
• Utilizing housing price data from 296 Chinese cities.
• Constructing a spatiotemporal correlation matrix and applying the directed minimum spanning tree algorithm.

Main Results:

• The Chinese urban housing price spillover network displays a hierarchical structure with significant ripple effects.
• Eastern coastal cities and the national core city act as primary radiation hubs for price contagion.
• The East China sub-network holds a unique net spillover position, and regional sub-networks show heterogeneous evolution patterns.

Conclusions:

• The CBEDTE-GM model effectively maps directed housing price spillover pathways, revealing network hierarchies and dominant hubs.
• Regional housing markets exhibit distinct evolutionary trajectories, shifting towards centralized or multi-polar structures.
• Findings offer insights into the complex dynamics of urban housing price contagion in China.