This paper estimates the relationship between the housing sale prices in the housing markets using a multivariate GARCH and GJR-GARCH model. The data set includes monthly returns in apartments, detached houses, and townhouses from January 2004 to May 2018. The main results are as follows. First, in the analysis of apartment returns, asymmetric volatility is found in three regions: Seoul, Incheon, and Gyeonggi. The correlation coefficient between Seoul and Gyeonggi is higher than between Seoul and Incheon or Incheon and Gyeonggi. Second, in the analysis of detached house returns, the correlation coefficients are smaller than those of apartments, and the estimate of asymmetric volatility is not significant in Incheon. Third, the correlation coefficient between Seoul and Gyeonggi is highest in the townhouse market, whereas the correlation coefficient between Incheon and Gyeonggi is smallest in that market. This result implies that, regardless of the housing type, the correlation between Seoul and Gyeonggi is higher than others.
2. 선행연구 검토
4. 실증분석 결과