On the Application of the Cross-Correlations in the Chinese Fund Market: Descriptive Properties and Scaling Behaviors

Weibing Deng, Wei Li, Xu Cai, and Qiuping Wang

Contact the author: Please use for correspondence this email.
Submission date: 20. Jul. 2011
Pages: 15
published in: Advances in complex systems, 14 (2011) 1, p. 97-109 
DOI number (of the published article): 10.1142/S0219525911002871
Bibtex
with the following different title: On the application of the cross-correlation in the Chinese fund market : descriptive properties and scaling behaviors
Keywords and phrases: cross-correlation, Random matrix theory, power-law
Download full preprint: PDF (334 kB)

Abstract:
On the basis of the relative daily logarithmic returns of 88 different funds in the Chinese fund market (CFM) from June 2005 to October 2009, we construct the cross-correlation matrix of the CFM. It is shown that the logarithmic returns follow an exponential distribution, which is commonly shared by some emerging markets. We hereby analyze the statistical properties of the cross-correlation coefficients in different time periods, such as the distribution, the mean value, the standard deviation, the skewness and the kurtosis. By using the method of the scaled factorial moment, we observe the intermittence phenomenon in the distribution of the cross-correlation coefficients. Also by employing the random matrix theory (RMT), we find a few isolated large eigenvalues of the cross-correlation matrix, and the distribution of eigenvalues exhibits the power-law tails. Furthermore, we study the features of the correlation strength with a simple definition.