경제시계열을 이용한 대전시 아파트가격 영향에 관한 연구

A Study on the Impact of Apartment Price on Daejeon City by Using Economic Time Series

In analyzing the real estate market, advanced research using macroeconomic variables have been conducted; however, this research is limited in dealing with monetary and fiscal policy. They do not properly represent the phenomenon of differences between regions. Apart from variables of the fiscal and monetary policy, which were studied in advanced research, this study analyzes a relationship of impact on apartment sales price in Daejeon City by utilizing the economic sentiment index, an leading index in consumption and economy, and variables directly related to the price: residential mortgage, apartment transactions and number of unsold apartments. The results of analysis are as follows: First, a unit root test was conducted to secure stability in the economic time series data. Stability in materials are secured through the first difference for CD interest rates, log conversion for apartment transactions and log first difference for other variables. Second, the results of the Granger causality test indicates that apartment sales prices are influenced by CD interest rates and number of unsold apartments. Third, according to the Johansen system cointegration test, two or more variables have a linear relationship. As the analysis was conducted by the VEC (Vector Error Correction) model, a long-run equilibrium relationship was identified for apartment sales price. Lastly, analysis of impulse response and variance decomposition shows that all variables have an impact on apartment sales price in the long term. This study examined short-run and long-run equilibrium relationships between variables through the VEC model by applying direct variables for apartment sales price in Daejeon; however, it is required not only to collect data in the long term, but also to study the development of proxy variables for apartment sales price further with the aim of overcoming constraints for variables in the economic time series data and ensuring detailed analysis. This requirement needs to be achieved in the future.