Cointegration

The concept of cointegration in the context of time series analysis and econometrics goes back to the British Clive WJ Granger and Robert F. Engle Americans. A cointegration relationship exists when there is a long-term balance between two or more non-stationary ( integrated ) variables. In the short term there will be deviations from equilibrium; at least one of the variables adapts over time so that the long-run equilibrium is restored.

About the Engle - Granger representation theorem - can be a relationship between the concept of cointegration of variables and an error correction model to produce. This means that for every Kointegrationsmodell an error correction model exists that describes the short-term dynamics of the system.

The cointegration is in trend affected time series ( non-stationary time series) applied. It represents an alternative to a trend adjustment approximately by subtraction dar. trend adjustment is often suggested to avoid bill regressions. The disadvantage of this treatment of non-stationarity is that lost by purging information. Here the advantage of the work lies with level variables and cointegration. Long-term equilibrium relationships can be detected and analyzed.

Clive WJ Granger and Robert F. Engle have been awarded, among others, for their work on cointegration 2003 with the Prize in Economic Sciences the Bank of Sweden in memory of Alfred Nobel.