Random matrix theory analysis of cross-correlations in the US stock market: Evidence from Pearson’s correlation coefficient and detrended cross-correlation coefficient |
| |
Authors: | Gang-Jin Wang Chi Xie Shou Chen Jiao-Jiao Yang Ming-Yan Yang |
| |
Institution: | 1. College of Business Administration, Hunan University, Changsha 410082, China;2. Center of Finance and Investment Management, Hunan University, Changsha 410082, China |
| |
Abstract: | In this study, we first build two empirical cross-correlation matrices in the US stock market by two different methods, namely the Pearson’s correlation coefficient and the detrended cross-correlation coefficient (DCCA coefficient). Then, combining the two matrices with the method of random matrix theory (RMT), we mainly investigate the statistical properties of cross-correlations in the US stock market. We choose the daily closing prices of 462 constituent stocks of S&P 500 index as the research objects and select the sample data from January 3, 2005 to August 31, 2012. In the empirical analysis, we examine the statistical properties of cross-correlation coefficients, the distribution of eigenvalues, the distribution of eigenvector components, and the inverse participation ratio. From the two methods, we find some new results of the cross-correlations in the US stock market in our study, which are different from the conclusions reached by previous studies. The empirical cross-correlation matrices constructed by the DCCA coefficient show several interesting properties at different time scales in the US stock market, which are useful to the risk management and optimal portfolio selection, especially to the diversity of the asset portfolio. It will be an interesting and meaningful work to find the theoretical eigenvalue distribution of a completely random matrix R for the DCCA coefficient because it does not obey the Mar?enko–Pastur distribution. |
| |
Keywords: | Econophysics Random matrix theory Cross-correlations Detrended Cross-correlation coefficient Pearson&rsquo s correlation coefficient US stock market |
本文献已被 ScienceDirect 等数据库收录! |
|