Robustness of Sign Correlation in Market Network Analysis
Financial market can be modeled as network represented by a complete weighted graph. Different characteristics of this graph (minimum spanning tree, market graph, and others) give an important information on the network. In the present paper it is studied how the choice of measure of similarity between stocks influences the statistical errors in the calculation of network characteristics. It is shown that sign correlation is a robust measure of similarity in contrast with Pearson correlation widely used in market network analysis. This gives a possibility to get more precise information on stock market from observations.
KeywordsPearson Correlation Minimum Span Tree Sample Sign Heavy Tail Mixture Distribution
This work is partly supported by RF government grant, ag. 11.G34.31.0057 and RFFI grant 14-01-00807.
- 1.Anderson, T.W.: An Introducion to Multivariate Statistical Analysis, 3rd edn. Wiley-Interscience, New York (2003)Google Scholar
- 2.Bautin, G.A. Kalyagin, V.A., Koldanov, A.P.: Comparative analysis of two similarity measures for the market graph construction. Springer Proceedings in Mathematics and Statistics, vol. 59, pp. 29–41 (2013). http://link.springer.com/chapter/10.1007%2F978-1-4614-8588-9_3
- 5.Gupta, F.K. Varga, T., Bodnar, T.: Elliptically Contoured Models in Statistics and Portfolio Theory. Springer (2013). ISBN: 978-1-4614-8153-9Google Scholar
- 6.Kalyagin, V.A., Koldanov, A.P., Koldanov, P.A., Pardalos, P.M., Zamaraev, V.A.: Measures of uncertainty in market network analysis. (2013) arXiv:1311.2273Google Scholar
- 8.Mantegna, R.N., Stanley, H.E.: An Introduction to Econophysics: Correlations and Complexity in Finance. Cambridge University Press, Cambridge (2000)Google Scholar
- 10.Shiryaev, A.N.: Essential of Stochastic Finance: Facts, Models, Theory. Word Scientific, New Jersey (2003)Google Scholar