对于门限向量误差修正模型给出原假设为线性非同积关系的门限同积bootstrap检验.给出上述模型中未识别参数的最大似然估计;提出检验门限同积关系的SupLM检验及相应的渐近分布;并给出了SupLM检验统计量p值的bootstrap模拟方法,模拟实验验证了该方法的有效性,进而应用该方法检验了几组美国国库券收益率问存在明显的门限同积关系。
This paper develops a test for the presence of threshold cointegation in a threshold vector error-correction model with the linear no cointegration null hypothesis. We propose a simple algorithm to obtain maximum likelihood estimation of the complete threshold cointegration model. We adopt a SupLM type test; derive its null asymptotic distribution and present bootstrap approximation. Simulation evidence shows that bootstrap inference generates moderate size and power of the test. Our method is illustrated with used U.S. treasury yield curve rates.