This paper examines if there is a long -run relationship between the time series representing asset markets. Before testing cointegration, I check the stationarity of the time series by unit root test, and determine the appropriate length of lag with ...
This paper examines if there is a long -run relationship between the time series representing asset markets. Before testing cointegration, I check the stationarity of the time series by unit root test, and determine the appropriate length of lag with information criterion and partial autocorrelation. I provide the evidence real estate market and financial market are insulated or segmented from each other, but apartment market and stock market are cointegrated. With structural change of real estate market in 1994, real estate(apartment) market and financial market have one common cointegrating vector. When the CPI is included in the systems, there are no effect on the results. But after 1997, with the CPI the number of cointegrating vector increase in all time series combinations. The findings have implications for developing mixed asset portfolio.