The multivariate time series model you choose to describe your data depends on whether there are cointegrating relations among the response series. For more details, see Cointegration and Error Correction Analysis.
Learn about cointegrated time series and error correction models.
The Engle-Granger test for cointegration and its limitations.
Test the null hypothesis that there are no cointegrating relationships among the response series composing a multivariate model.
Learn about the Johansen test for cointegration.
Assess whether a multivariate time series has multiple cointegrating relations using the Johansen test.
Compare Johansen and Engle-Granger approaches to cointegration analysis.
Learn testing linear constraints on cointegrating relations and adjustment speeds about using the Johansen framework.
Conduct tests on cointegrating vectors.
Conduct tests on adjustment speeds.
Compute and interpret the cointegration rank of a VEC model.
Estimate the parameters of a VEC model.
Generate forecasts from a VEC model.
Generate impulse responses from a VEC model.
Produce maximum likelihood estimates of VEC model coefficients under the rank restrictions on the cointegrating matrix.