Table 3.
Johansen and Juselius co-integration tests.
| Variables | Null | Optimal lag | Test statistics |
|
|---|---|---|---|---|
| Trace statistics | Maximum eigenvalue | |||
| LELECSA and LGDPSA#2 | r = 0 | 1 | 22.95* | 19.89* |
| r ≤ 1 | 3.06 | 3.06 | ||
| LELECSA, LGDPSA and LPOP#2 | r = 0 | 1 | 28.98 | 17.61 |
| r ≤ 1 | 11.36 | 7.99 | ||
| r ≤ 2 | 3.37 | 3.37 | ||
| LELECSA, LGDPSA and LTOURSA#2 | r = 0 | 1 | 36.54* | 23.82* |
| r ≤ 1 | 12.72 | 9.03 | ||
| r ≤ 2 | 3.69 | 3.69 | ||
| LELECSA, LGDPSA, LPOP and LTOURSA#2 | r = 0 | 1 | 50.90 | 28.44 |
| r ≤ 1 | 22.46 | 11.44 | ||
| r ≤ 2 | 11.01 | 6.18 | ||
| r ≤ 3 | 4.83 | 4.83 | ||
Notes: The optimal lag lengths are chosen using Schwartz Information Criteria.
* denotes the rejection of the null hypothesis at the 5 percent significance value.
# denotes the co-integration method used as follows: #1. No intercept or trend in co-integrating equation (CE) or VAR, #2. Intercept (no trend) in CE, no intercept or trend in VAR, #3. Intercept in CE and VAR, no trends in CE and VAR, #4. Intercept in CE and VAR, linear trend in CE, no trend in VAR, and #5. Intercept and quadratic trend in the CE and intercept and linear trend in VAR.