Table 3.
CIPS test
| H0: all series within the panel are nonstationary Ha: some series within the panel are stationary |
||||
|---|---|---|---|---|
| Levels | First differences | |||
|
Zt-bar (1) |
p-value (2) |
Zt-bar (3) |
p-value (4) |
|
| INV | 0.106 | 0.542 | –6.531 | 0.000 |
| TGE / GDP | -0.418 | 0.338 | –10.939 | 0.000 |
| PGE / GDP | 1.040 | 0.851 | –2.571 | 0.005 |
| UGE / GDP | 1.880 | 0.970 | –3.313 | 0.000 |
| PGE / TGE | -2.490 | 0.006 | –9.992 | 0.000 |
| UGE / TGE | -2.465 | 0.007 | –9.558 | 0.000 |
| TAX | -0.472 | 0.319 | –10.325 | 0.000 |
| CAPB | -2.317 | 0.010 | –8.464 | 0.000 |
| GDP | 0.926 | 0.823 | –3.536 | 0.000 |
| INTEREST | -3.704 | 0.000 | –10.181 | 0.000 |
Notes: The optimal lag length is selected according to the Akaike Information Criterion (AIC). The intercept and linear time trends are included in the equations. Columns 1 and 3 report the test statistics, whereas the related p-values are reported in columns 2 and 4. The first (second) two columns refer to the model in levels (first differences)