A Study on the Impact of the International Exchange Complex District Development on Surrounding Real Estate: Focused on the Third General Residential Area of Gangnam-gu

This study empirically analyzes the impact of the International Exchange Complex Zone development in Seoul's Gangnam on land and building unit prices, focusing on the significance of spatial autocorrelation. Utilizing 1,339 actual transaction cases from 2006 to 2024, OLS, Spatial Lag Model (SAR), and Spatial Error Model (SEM) were comparatively analyzed. OLS regression results indicated that development phases, property area, macroeconomic indicators, and location variables significantly influenced both dependent variables. However, OLS residual diagnostics revealed statistically significant spatial autocorrelation, underscoring the necessity of applying spatial econometric models and leading to a 1000m optimal spatial threshold selection. SAR and SEM models, applied with this threshold, demonstrated comparable or improved explanatory power (R²/Pseudo R²) compared to OLS, with SEM showing the best fit by information criteria (AIC, BIC). Moran's I tests on their residuals confirmed successful resolution of spatial autocorrelation. Significant spatial coefficients (in SAR and SEM) proved strong spatial spillover effects and the influence of unobserved spatial characteristics, highlighting OLS estimation bias. This research demonstrates that spatial autocorrelation significantly impacts real estate price determination and that spatial econometric models are essential for reliable analysis and spatially informed policymaking. However, it acknowledges limitations in estimating pure causal effects, suggesting future research incorporate causal inference methodologies