Affordable Access

Publisher Website

Hidden cross-correlation patterns in stock markets based on permutation cross-sample entropy and PCA

Physica A Statistical Mechanics and its Applications
Publication Date
DOI: 10.1016/j.physa.2014.08.064
  • Permutation
  • Cross-Sample Entropy
  • Stock Market
  • Principle Component Analysis (Pca)
  • Economics


Abstract In this article, we investigate the hidden cross-correlation structures in Chinese stock markets and US stock markets by performing PCSE combined with PCA approach. It is suggested that PCSE can provide a more faithful and more interpretable description of the dynamic mechanism between time series than cross-correlation matrix. We show that this new technique can be adapted to observe stock markets especially during financial crisis. In order to identify and compare the interactions and structures of stock markets during financial crisis, as well as in normal periods, all the samples are divided into four sub-periods. The results imply that the cross-correlations between Chinese group are stronger than the US group in the most sub-periods. In particular, it is likely that the US stock markets are more integrated with each other during global financial crisis than during Asian financial crisis. However, our results illustrate that Chinese stock markets are not immune from the global financial crisis, although less integrated with other markets if they are compared with US stock markets.

There are no comments yet on this publication. Be the first to share your thoughts.


Seen <100 times

More articles like this

Hidden cross-correlation patterns in stock markets...

on Physica A Statistical Mechanic...

Forbidden patterns, permutation entropy and stock...

on Physica A Statistical Mechanic... Jan 01, 2009

A permutation entropy based test for causality: Th...

on Physica A Statistical Mechanic... Mar 15, 2014

Volatility and cross correlation across major stoc...

on Journal of Empirical Finance Jan 01, 1998
More articles like this..