Long-range correlation and market segmentation in bond market

Zhongxing Wang,Yan Yan,Xiaosong Chen

DOI: https://doi.org/10.1016/j.physa.2017.04.066

2017-09-01

Abstract:This paper investigates the long-range auto-correlations and cross-correlations in bond market. Based on Detrended Moving Average (DMA) method, empirical results present a clear evidence of long-range persistence that exists in one year scale. The degree of long-range correlation related to maturities has an upward tendency with a peak in short term. These findings confirm the expectations of fractal market hypothesis (FMH). Furthermore, we have developed a method based on a complex network to study the long-range cross-correlation structure and applied it to our data, and found a clear pattern of market segmentation in the long run. We also detected the nature of long-range correlation in the sub-period 2007–2012 and 2011–2016. The result from our research shows that long-range auto-correlations are decreasing in the recent years while long-range cross-correlations are strengthening.

physics, multidisciplinary