摘要：This paper introduces a spectral clustering-based method to show that stock prices containnot only firm but also network-level information. We cluster different stock indices and reconstructthe equity index graph from historical daily closing prices. We show that tail events have a minoreffect on the equity index structure. Moreover, covariance and Shannon entropy do not provideenough information about the network. However, Gaussian clusters can explain a substantial part ofthe total variance. In addition, cluster-wise regressions provide significant and stationer results.
关键词：cluster analysis; equity index networks; machine learning
