
Figure 2. The logs of seasonally adjusted money supply and exchange rate
5.60
5.55
5.50
5.45
5.40
5.35
5.30

97 98 99 00 01 02 03 04
I — Lnxreuro |
4.1. Stationarity and integration tests
First, the Elliott, Rothenberg, Stock (1996) DF-GLS unit root test, with and without a linear
trend is performed. The results are presented in Table 1.
Table 1. DF-GLS unit root tests
Variables |
Specification |
Lags |
Test statistic |
lnIPI |
constant |
5 |
- 0.904 |
constant and trend |
5 |
- 2.997 | |
lnPPI |
constant |
3 |
- 1.722 |
constant and trend |
3 |
- 2.349 | |
lnM1A |
constant |
5 |
0.366 |
constant and trend |
5 |
- 0.697 | |
InXREURO |
constant |
2 |
-0.264 |
constant and trend |
2 |
- 0.931 |
The critical values for 0.95 (0.99) confidence levels with constant are -1.944 (-2.592), with constant and trend
are -3.074 (-3.633). The Akaike Information Criteria was used to determine the lag length.
None of the tests statistics is significant, all the variables appears to be integrated. To ensure
that all series are I(1), and not integrated of a higher order, the first differences are tested
using the DF-GLS unit root tests in table 2. Because there is no evidence of a linear trend in
the first difference representation of the variables, we conduct the second order unit root tests
More intriguing information
1. DIVERSITY OF RURAL PLACES - TEXAS2. Land Police in Mozambique: Future Perspectives
3. THE WAEA -- WHICH NICHE IN THE PROFESSION?
4. The name is absent
5. The name is absent
6. BODY LANGUAGE IS OF PARTICULAR IMPORTANCE IN LARGE GROUPS
7. Applications of Evolutionary Economic Geography
8. The name is absent
9. Naïve Bayes vs. Decision Trees vs. Neural Networks in the Classification of Training Web Pages
10. The storage and use of newborn babies’ blood spot cards: a public consultation