마르코프 국면전환모형을 이용한 부동산 경기변동 분석

This paper is to model changes in regimes by using Seoul apartment sale price indices from December, 2002 to October, 2010. The parameters of an autoregression are viewed as the outcome of a discrete-state Markov process, which the mean growth rate of a nonstationary series may be subject to occasional, discrete shifts. An empirical application of this technique to apartment sales price indices in Seoul suggests that the periodic shift from a positive growth rate to a negative growth rate is a recurrent feature of apartment sales business cycle. The estimated parameter values suggest that positive and negative growth rates differ across areas and time, implying that the government should take into account in what regime real estate business cycles are when it decides real estate policies. Most areas show large volitilities of real estate business cycles and shorter durations in expansion regimes than in contraction ones. The results also shows the possibility of forecasting the regime switching in the future which is of much interest for real estate market participants.