Cointegrationsbeziehung

Cointegrationsbeziehung is a statistical concept in econometrics that describes the long-term relationship between two or more variables that are both integrated of the same order. In other words, it refers to the state of being cointegrated, where two or more time series that are individually non-stationary become stationary when combined in a specific way.

The concept of cointegration was introduced by Robert F. Stambaugh in 1986 and is often used in conjunction with other econometric techniques, such as error correction models. Cointegration allows researchers to identify the long-term relationships between variables that cannot be explained by short-term fluctuations.

A cointegrationsbeziehung exists when at least two variables are integrated of the same order, typically I(1) or I(2), and can be combined to form a stationary combination, also known as a cointegrating relationship. This stationarity is achieved when the sum of the individual time series of the variables is stationary, which means that their long-term movements are correlated.

The cointegrationsbeziehung is often used in various economic applications, such as testing the Purchasing Power Parity (PPP) hypothesis, analyzing the relationship between stock prices and inflation, or examining the long-term relationships between exchange rates and macroeconomic variables.

Estimating a cointegrationsbeziehung typically involves a Johansen test, which examines the number of cointegrating relationships between variables, as well as vector autoregressions (VARs) or error correction models, which are used to estimate the coefficients of the cointegrating relationships.