Nonparametric cointegration analysis



with (the log of) real money, the real interest rate and the inflation rate. It seems that only
three out of seventeen tested series are likely (close to) genuine unit root processes, namely
the logs of GNP, wages and velocity of money. For the other series the test results were
either in favor of the trend stationarity hypothesis or inconclusive, in the sense that not all the
tests favored the same hypotheses. The latter results may be due to the presence of trend
breaks, but we did not test for that. Next, inspection of the plots of the remaining three series
(see Schotman and Van Dijk, 1991) revealed possible cointegration of the logs of GNP and
wages. Therefore, we selected these two series for our cointegration analysis. The two-
dimensional vector time series involved has length
n = 80 (from 1909 to 1988). Finally, as a
double check we applied our nonparametric cointegration test to each of the two series: the
unit root hypothesis could not be rejected at the 10% significance level.

8.2. Nonparametric cointegration analysis

The result of our nonparametric cointegration analysis is that the null hypothesis of no
cointegration (
r = 0) is rejected at the 5% significance level, whereas the null hypothesis r =
1 is accepted at the 10% significance level. Thus we conclude that ln(wages) and ln(GNP) are
cointegrated:
r = 1. This result is confirmed by the estimation approach in section 4.4: the
function
gm(r) defined by (26), with m = 2, takes the values gm(0) = 1382.966, gm(1) = 3.087,
gm(2) = 28164.158, hence the estimated number of cointegrated vectors is 1. The estimate of
the standardized cointegrating vector is (1, -.70)
T, i.e., ln(wages) - .7ln(GNP) is (trend)
stationary.

In order to see how "significant" the estimated cointegrating vector is, we have
conducted a series of trace tests (which in this case coincide with the lambda-max tests), for 2
× 1 matrices
H = (1,a)T with

a {-.4,-.5,-.6.-.65,-.7,-.75,-.8,-.9,-1}.

The null hypothesis is accepted at the 10% significance level for a ranging from -.6 to -.8,
and at the 5% level for
a ranging from -.5 to -.9.

8.3. Johansen’s approach

26



More intriguing information

1. Behavioural Characteristics and Financial Distress
2. Cyclical Changes in Short-Run Earnings Mobility in Canada, 1982-1996
3. The name is absent
4. La mobilité de la main-d'œuvre en Europe : le rôle des caractéristiques individuelles et de l'hétérogénéité entre pays
5. The quick and the dead: when reaction beats intention
6. El Mercosur y la integración económica global
7. LOCAL PROGRAMS AND ACTIVITIES TO HELP FARM PEOPLE ADJUST
8. Large-N and Large-T Properties of Panel Data Estimators and the Hausman Test
9. CGE modelling of the resources boom in Indonesia and Australia using TERM
10. Learning and Endogenous Business Cycles in a Standard Growth Model
11. THE WAEA -- WHICH NICHE IN THE PROFESSION?
12. Infrastructure Investment in Network Industries: The Role of Incentive Regulation and Regulatory Independence
13. Life is an Adventure! An agent-based reconciliation of narrative and scientific worldviews
14. The name is absent
15. Analyse des verbraucherorientierten Qualitätsurteils mittels assoziativer Verfahren am Beispiel von Schweinefleisch und Kartoffeln
16. Categorial Grammar and Discourse
17. The name is absent
18. Migration and Technological Change in Rural Households: Complements or Substitutes?
19. Centre for Longitudinal Studies
20. The name is absent