Eviews9.0中EG两步法结果怎么看? Cointegration Test - Engle-

玲。。 |浏览3678次
2019/03/21 20:12
Eviews9.0中EG两步法结果怎么看? Cointegration Test - Engle-Granger Date: 03/21/19 Time: 19:52 Equation: UNTITLED Specification: LNGDP LNKPAT C Cointegrating equation deterministics: C Null hypothesis: Series are not cointegrated Automatic lag specification (lag=0 based on Schwarz Info Criterion, maxlag=1) Value Prob.* Engle-Granger tau-statistic -3.107177 0.1931 Engle-Granger z-statistic -10.27508 0.0396 *MacKinnon (1996) p-values. Warning: p-values may not be accurate for fewer than 20 observations. Intermediate Results: Rho - 1 -1.467869 Rho S.E. 0.472412 Residual variance 0.000231 Long-run residual variance 0.000231 Number of lags 0 Number of observations 7 Number of stochastic trends** 2 **Number of stochastic trends in asymptotic distribution. Engle-Granger Test Equation: Dependent Variable: D(RESID) Method: Least Squares Date: 03/21/19 Time: 19:52 Sample (adjusted): 2011 2017 Included observations: 7 after adjustments Variable Coefficient Std. Error t-Statistic Prob. RESID(-1) -1.467869 0.472412 -3.107177 0.0209 R-squared 0.604328 Mean dependent var -0.004022 Adjusted R-squared 0.604328 S.D. dependent var 0.024155 S.E. of regression 0.015194 Akaike info criterion -5.404278 Sum squared resid 0.001385 Schwarz criterion -5.412005 Log likelihood 19.91497 Hannan-Quinn criter. -5.499784 Durbin-Watson stat 1.828885
  • 问题补充 : 2019/03/21 20:19

    这是做协整。主要有两个问题。1、我做的这两个变量,它的原序列是平稳的,但是我要做格兰杰因果检验,是直接做格兰杰应该检验?还是用EG两步法做协整然后再做格兰杰因果检验?2、这个结果是用EG两步法做协整得到的,不是像传统的那样先做OLS回归然后对残差进行ADF检验得到的,是直接做回归然后用EG得到的,但是从这个结果中如何看残差协不协整?

收藏关注
全部回答(1)
0人关注该问题
 加载中...