kointegrerende

Kointegrerende is a term used in econometrics and time series analysis. It describes a situation where two or more non-stationary time series are linearly related in such a way that a linear combination of them is stationary. Non-stationary time series are those whose statistical properties, like mean and variance, change over time. Often, these series exhibit a unit root, meaning they tend to drift away from their mean.

When a set of time series is cointegrated, it implies that they share a common underlying long-run trend. Although individual series may wander unpredictably in the short term, they tend to move back towards a stable equilibrium relationship. This long-run relationship is captured by the cointegrating vector, which represents the coefficients of the linear combination.

The concept of cointegration is important because it allows for the modeling of long-run equilibrium relationships between economic variables that might appear to be unrelated in the short run. If series are cointegrated, a model that accounts for this long-run relationship, such as an Error Correction Model (ECM), can be more effective than one that treats the series as independent. Testing for cointegration typically involves statistical tests like the Engle-Granger test or the Johansen test. The presence of cointegration suggests that deviations from the long-run equilibrium are temporary and will be corrected over time.