Sigma Plus Consulting
 

Cointegration

Many market practitioners base their analysis of the relationships within multivariate data upon the concept of correlation. Correlation is a simple linear relationship between two series of values, and will often oversimplify the complexity of interaction between two or more time series. Cointegration is a more powerful technique for investigating common trends in multivariate time series.

Cointegration begins by looking for long term relationships between asset prices, and then using these relationships for a dynamic analysis of correlation of returns. Correlation based techniques look only at returns, while cointegration techniques use base price and return data.

The graph above shows two cointegrated exchange rates.

Cointegration models can be used for

For further information contact us at info@sigmaplus.com.au

Send mail to webmaster@sigmaplus.com.au with questions or comments about this web site.
Last modified: November 16, 2002