Table 4: Russia: Johansen cointegration test; with linear deterministic trend and lag interval (in
first differences) 1 to 2
Unrestricted Cointegration Rank Tests
|
Hypothesized |
Eigen-value |
Trace Statistic |
5 Percent Critical Value |
1 percent Critical Value |
|
None |
0.524723 |
122.95190 |
87.31 |
96.58 |
|
At most 1 |
0.371151 |
62.69944 |
62.99 |
70.05 |
|
At most 2 |
0.139696 |
25.12649 |
42.44 |
48.45 |
|
At most 3 |
0.100898 |
12.93849 |
25.32 |
30.45 |
|
At most 4 |
0.051976 |
4.32339 |
12.25 |
16.26 |
|
Hypothesized |
Eigen-value |
Max-Eigen |
5 Percent Critical Value |
1 percent |
|
None |
0.524723 |
60.25245 |
37.52 |
42.36 |
|
At most 1 |
0.371151 |
37.57296 |
31.46 |
36.65 |
|
At most 2 |
0.139696 |
12.18800 |
25.54 |
30.34 |
|
At most 3 |
0.100898 |
8.61509 |
18.96 |
23.65 |
|
At most 4 |
0.051976 |
4.32339 |
12.25 |
16.26 |
In Tables 5 and 6, the estimated cointegrating equations for the logarithm of real exchange
rate for Poland and Russia are presented. The t-statistics are in parentheses. Specification 1 in
each table is the initial specification. In order to correct for possible serial correlation and
simultaneity bias the equations are re-specified with DOLS. Specification 2 is the final result. To
confirm that the vector of the estimated coefficients is a cointegrating vector, ADF and Phillips-
Perron unit root tests are conducted on the residual series. The relevant critical values are
obtained from Philips and Ouliaris (1990, Table IIc). The tests strongly reject the null hypothesis
23