Cluster-Based Investigation of Stock Market Indices

We introduce a spectral clustering-based method to show that stock prices contain not only firm, but also network level information. We cluster different stock indices and reconstruct the equity index graph from historical daily closing prices. We find that tail events have a minor effect on the equity index structure. Gaussian clusters can explain a substantial part of the total variance. Thus, mean-variance analysis with Gaussian clusters gives significant regression estimations. In addition, cluster-wise regressions also provide significant and stationer results.