最强美国论文代写推荐：test method
最强美国论文代写推荐：test method
恩格尔和格兰杰（1987）发现了一种测试统计学中序列之间长期均衡的方法。他们的方法称为Engel-Granger两步法。开发了Engel-Granger两步程序来测试零假设，即在一对I（1）系列中不存在协整关系。他们使用普通最小二乘法估计这些变量之间的固定关系，然后使用单位根检验来检验残差。拒绝单位根零假设的存在是协整关系存在的证据。
Johansen（1988）提出了另一种协整检验方法，他开发了一种多变量协整检验的最大似然估计方法。该方法可以测试协整向量的线性约束，解决了对称性和比例条件的测试精度问题。此过程非常复杂，但可以通过某些统计分析软件包（例如，Eviews7）轻松获得结果。进一步的细节可以在Johansen和Juselius（1990）和Johansen（1991）中找到。与三变量系统相比，双变量系统使用更多的协整技术。这两种方法的主要区别在于Engel-Granger两步法使用单方程技术，而Johansen协整检验使用多方程技术。因此，Johansen协同测试在应用领域的假设中具有较少的限制。
最强美国论文代写推荐：test method
Engel and Granger (1987) found a method to test the long-run equilibrium between series in statistics. Their method is called Engel-Granger two-step procedure. Engel-Granger two-step procedure is developed to test the null hypothesis that there’s no co-integration relationships existing in a pair of I (1) series. They used ordinary least squares method to estimate a stationary relationship between these variables, then use the unit root test to test the residuals. Rejecting the existence of unit root null hypothesis is the evidence of the existence of a co-integration relationship.
An alternative approach of co-integration test is proposed by Johansen (1988), who developed a maximum likelihood estimation method for multivariate co-integration tests. The method can test the linear constraints of co-integrating vectors, and has solved the problem of test accuracy of symmetry and proportionality conditions. This procedure is very complex, but the result can be easily obtained by some Statistical analysis software packages (for instance, Eviews7). Further details can be found in Johansen and Juselius (1990) and Johansen (1991). Compared to three-variable system, two-variable system uses more co-integration techniques.
The main difference between these two methods is that the Engel-Granger two-step procedure uses the single-equation techniques, and the Johansen co-integration test uses a multi-equation technique. So Johansen co-integration test has fewer limitations in the assumptions the application fields.
