Explanation:
When two time series each have a unit root (are non-stationary) but are not cointegrated, this means there is no long-run equilibrium relationship between them.
Key implications:
Therefore, both the error term non-stationarity and inconsistent coefficients occur when non-stationary series are not cointegrated.
An analyst regresses changes in an exchange rate over time on the two countries' inflation rate differential. While each of the two time series has a unit root, the time series are not cointegrated. This implies that the error term in the linear regression will:
A
be covariance stationary and the regression coefficients will be consistent.
B
be covariance stationary but the regression coefficients will not be consistent.
C
not be covariance stationary and the regression coefficients will not be consistent.
No comments yet.