Journal of Network Theory in Finance

Interconnectedness risk and active portfolio management: the information-theoretic perspective

Eduard Baitinger and Jochen Papenbrock

  • Correlation networks are not appropriate for interconnectedness risk quantification if the dependence structure is non-linear.
  • Mutual information networks provide an advantage when non-linear dependencies exist, without losing effectiveness given linear only dependencies. 
  • Active investment strategies based on mutual information networks outperform conventional strategies relying on linear dependence measures.

Today’s asset management academia and practice are dominated by mean–variance thinking. Consequently, this usually leads to the quantification of the dependence structure of asset returns by the covariance or the Pearson correlation coefficient matrix. The respective dependence measures are linear by construction and hence unable to detect nonlinear dependencies. This paper tackles the described concern with regard to financial networks and their implementation in active investment strategies. We discuss the mutual information measure, which is an information-theoretic concept and is able to detect linear and nonlinear dependencies. The empirical part of this paper extensively compares mutual-information-based networks with correlation-based networks on a stand-alone basis and in the framework of active investment strategies.

Sorry, our subscription options are not loading right now

Please try again later. Get in touch with our customer services team if this issue persists.

New to View our subscription options

You need to sign in to use this feature. If you don’t have a account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here