Cointegrating regression on cyclical dataset

[broncus2k]

Hi,


From what I know, cointegrating regression (COINTREG method in eveiws) is used when both the dependent and independent variables are I(1) processes.

Under these circumstances, cointegrating regression is preferred over OLS-type specification because it is able to detect spurious regressions.


My question is, when we are dealing with dependent and independent variables that exhibit cyclical behaviour (which isn't I(1) process by the way), should we consider using cointegrating regression?


Thank you :)

[startz]

No. Cointegration only applies when both variables are integrated.