Table Summary of results from the ADF test
|
Test Number
|
Oil
|
GDP
|
Interest rate
|
Inflation
|
Unemployment
|
Exchange Rate
|
|
1
|
1.169921
|
-3.4819***
|
-2.35867**
|
-2.23904**
|
-0.98162
|
-1.05852
|
|
2
|
0.16054
|
-6.0342***
|
-1.7243***
|
-3.5581***
|
-2.15841
|
-2.40748
|
|
3
|
-3.72037*
|
-6.0839***
|
-2.795304
|
-3.62005**
|
-1.70094
|
-2.67032
|
***/**/* - Denotes that null hypothesis can be rejected at the 1%/5%/10% significance levels respectively.Figures condensed from Appendix tables 1A -> 1F.
From Table it can be seen that the variables; Oil, GDP, interest rate and inflation are all stationary and can reject the null hypothesis at the normal significance levels. The critical values for this test are those derived by James Mackinnon (1996). Therefore they do not contain unit roots. However for the remaining two variables -exchange rates and unemployment - the null hypothesis cannot be rejected at the normal significance levels.Therefore it is confirmed that these variables are not stationary and do contain a unit root. According to Kennedy (2001) this means that the regressions for these variables will be spurious meaning that these regressions are likely to show a very high value, indicating a strong goodness of fit. Also it might show t-statistics which deem that the coefficients are significant. However, these results may not have any economic significance at all.