Trouble Interpreting Johansen Cointegration
Posted: Tue Jul 05, 2011 8:46 am
Hello...
I'm an undergraduate student and now i'm writing my undergraduate thesis, it's about tea market market efficiency in grade level and my advisor suggest to use VAR model to analyze it, but i have difficulties in interpreting the Johansen Cointegration Test (Because i haven't studied about VAR and Economectric till now, i only studied basic statictics), So I need a little help in interpreting the result, the output from the test show like this;
Date: 06/13/11 Time: 22:26
Sample (adjusted): 3 94
Included observations: 92 after adjustments
Trend assumption: Linear deterministic trend
Series: LOG(JTA) LOG(CTA) LOG(MTA)
Lags interval (in first differences): 1 to 1
Unrestricted Cointegration Rank Test (Trace)
Hypothesized Trace 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None 0.148983 29.76459 29.79707 0.0504
At most 1 0.105376 14.92283 15.49471 0.0608
At most 2 * 0.049582 4.678485 3.841466 0.0305
Trace test indicates no cointegration at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
The test indicated no cointegration, but at 'At most 2' the Probability (P-value) is lower than 0.05. Should I use VECM, or should I continue with VAR Model based on the result ?
Any help would be appriciated
Thanks
I'm an undergraduate student and now i'm writing my undergraduate thesis, it's about tea market market efficiency in grade level and my advisor suggest to use VAR model to analyze it, but i have difficulties in interpreting the Johansen Cointegration Test (Because i haven't studied about VAR and Economectric till now, i only studied basic statictics), So I need a little help in interpreting the result, the output from the test show like this;
Date: 06/13/11 Time: 22:26
Sample (adjusted): 3 94
Included observations: 92 after adjustments
Trend assumption: Linear deterministic trend
Series: LOG(JTA) LOG(CTA) LOG(MTA)
Lags interval (in first differences): 1 to 1
Unrestricted Cointegration Rank Test (Trace)
Hypothesized Trace 0.05
No. of CE(s) Eigenvalue Statistic Critical Value Prob.**
None 0.148983 29.76459 29.79707 0.0504
At most 1 0.105376 14.92283 15.49471 0.0608
At most 2 * 0.049582 4.678485 3.841466 0.0305
Trace test indicates no cointegration at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
The test indicated no cointegration, but at 'At most 2' the Probability (P-value) is lower than 0.05. Should I use VECM, or should I continue with VAR Model based on the result ?
Any help would be appriciated
Thanks