Using Nonlinear Tests to Examine Integration Between Real Estate and Stock Markets
提出一种允许随机趋势项的非线性协整检验方法，用于分析房地产与股票市场的关系。结果表明传统线性检验支持市场分割，而非线性模型支持分数整合，两市场间存在缓慢调整的非线性关系。
An alternative approach to test whether the real estate and stock markets are cointegrated is presented. A nonlinear test, which allows for a stochastic trend term as opposed to a deterministic drift term, is developed. The results of the nonlinear model are compared to the results obtained using conventional cointegration tests. The cointegration results support the view that the real estate and stock markets are segmented, whereas the nonlinear model supports the view that the markets are fractionally integrated. There is a nonlinear relationship between the stock and real estate markets, but movement of the real estate market towards the stock market is slow and divergence between the two markets can be prolonged.