258
G. Marti et al.
10.5.6 Based on Several Corporate Networks
• Reference [34] combines different types of networks (ownership links, social ties
through board members, research collaborations, and stock correlations) into a
single multiplex structure. Authors employs several methods to show the significance of this multilayer network. Their initial results indicate a relation between
company performance and multiplex centrality.
10.6 Financial Applications
Though many of the academic studies focus on the MST or the clusters per se, some
papers try to extend their use beyond the filtering of empirical correlation matrices. It
has been proposed to leverage them for making financial policies, optimizing portfolios, computing alternative Value-at-Risk measures, residualizing expected returns,
grouping and selecting quantitative trading alphas, etc.
10.6.1 Portfolio Design
• Reference [108] finds that the Markowitz portfolio layer in the MST is higher than
the mean layer at all times.
• As the stocks of the minimum risk portfolio are found on the outskirts of the tree
[108, 119], authors expect larger trees to have greater diversification potential.
• In [114, 139], authors compare the Markowitz portfolios from the filtered empirical
correlation matrices using the clustering approach, the RMT approach and the
shrinkage approach.
• References [116, 122] propose to invest in different part of the MST depending
on the estimated market conditions.
• Authors show that there is no inner-mathematical relationship between the minimum variance portfolio from Markowitz theory and the portfolios designed from
the minimum spanning tree [64]. Empirical evidence of such relations found by
previous studies is essentially a stylized fact of financial returns correlations and
time series, not a general property of correlation matrices.
• It appears that a large number of stocks are unnecessary for building an index of
market change [74].
• The paper [38] describes methods for index tracking and enhanced index tracking
based on clusters of financial time series.
• Reference [40] introduces a procedure to design portfolios which are diversified
in their tail behavior by selecting only a single asset in each cluster.
G. Marti et al.
10.5.6 Based on Several Corporate Networks
• Reference [34] combines different types of networks (ownership links, social ties
through board members, research collaborations, and stock correlations) into a
single multiplex structure. Authors employs several methods to show the significance of this multilayer network. Their initial results indicate a relation between
company performance and multiplex centrality.
10.6 Financial Applications
Though many of the academic studies focus on the MST or the clusters per se, some
papers try to extend their use beyond the filtering of empirical correlation matrices. It
has been proposed to leverage them for making financial policies, optimizing portfolios, computing alternative Value-at-Risk measures, residualizing expected returns,
grouping and selecting quantitative trading alphas, etc.
10.6.1 Portfolio Design
• Reference [108] finds that the Markowitz portfolio layer in the MST is higher than
the mean layer at all times.
• As the stocks of the minimum risk portfolio are found on the outskirts of the tree
[108, 119], authors expect larger trees to have greater diversification potential.
• In [114, 139], authors compare the Markowitz portfolios from the filtered empirical
correlation matrices using the clustering approach, the RMT approach and the
shrinkage approach.
• References [116, 122] propose to invest in different part of the MST depending
on the estimated market conditions.
• Authors show that there is no inner-mathematical relationship between the minimum variance portfolio from Markowitz theory and the portfolios designed from
the minimum spanning tree [64]. Empirical evidence of such relations found by
previous studies is essentially a stylized fact of financial returns correlations and
time series, not a general property of correlation matrices.
• It appears that a large number of stocks are unnecessary for building an index of
market change [74].
• The paper [38] describes methods for index tracking and enhanced index tracking
based on clusters of financial time series.
• Reference [40] introduces a procedure to design portfolios which are diversified
in their tail behavior by selecting only a single asset in each cluster.