Journal of Network Theory in Finance

Risk.net

Visibility graph combined with information theory: an estimator of stock market efficiency

Bruna Amin Gonçalves and A. P. F. Atman

  • Emerging countries have on average the larger values of the Shannon entropy and lower values of the Fisher Information measure in Shannon-Fisher plane by visibility graph. 
  • Developed countries have on average the lower values of the Shannon entropy and larger values of the Fisher Information measure in Shannon-Fisher plane by visibility graph.  
  • It was possible to verify that in the period between January 2008 to May 2016, significant changes caused by Europe's crisis in 2008 have impacted the persistence features in these markets.

The visibility graph (VG) is a technique that acts as a bridge between dynamic systems and graph theory, and has been applied in recent years to analyze different systems in an innovative way. In this paper, we use information theory quantifiers to analyze the graphs generated by the VG method as applied to the return rate time series of stock markets from different countries between 1995 and 2016. In particular, we consider the degree probability distribution of the generated graphs to calculate Shannon entropy and Fisher information, in order to build the Shannon–Fisher plane. By analyzing the pattern of the countries along this plot, we demonstrate by this methodology that, on average, developed countries are concentrated at the top right-hand side of the plane, while emerging countries appear mostly at the bottom of the plane; we also identify markets that form a community. Moreover, when a separate analysis is applied to data considering two periods (before and after 2008), it is possible to detect significant changes in the patterns, mainly for developed countries; this indicates a possible quantifier for the impact caused by the 2008 European crisis in the global markets.

To continue reading...

You need to sign in to use this feature. If you don’t have a Risk.net 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 indvidual account here: