Mapping Coupled Timeseries Onto Complex Network
Abstract
In order to extract hidden joint information from two possibly uncorrelated timeseries, we explored the measures of network science. Alongside common methods in timeseries analysis of the economic markets, mapping the joint structure of two timeseries onto a network provides insight into hidden aspects embedded in the couplings. We discretize the amplitude of two timeseries and investigate relative simultaneous locations of those amplitudes. Each segment of a discretized amplitude is considered as a node. The simultaneity of the amplitudes of the two timeseries is considered as the edges in the network. The frequency of occurrences forms the weighted edges. In order to extract information, we need to measure that to what extent the coupling deviates from the coupling of two uncoupled series. Also, we need to measure that to what extent the couplings inherit their characteristics from a Gaussian distribution or a nonGaussian distribution. We mapped the network from two surrogate timeseries. The results show that the couplings of markets possess some features which diverge from the same features of the network mapped from white noise, and from the network mapped from two surrogate timeseries. These deviations prove that there exist joint information and crosscorrelation therein. By applying the network's topological and statistical measures and the deformation ratio in the joint probability distribution, we distinguished basic structures of crosscorrelation and coupling of crossmarkets. It was discovered that even two possibly known uncorrelated markets may possess some joint patterns with each other. Thereby, those markets should be examined as coupled and \textit{weakly} coupled markets.
 Publication:

arXiv eprints
 Pub Date:
 April 2020
 arXiv:
 arXiv:2004.13536
 Bibcode:
 2020arXiv200413536A
 Keywords:

 Quantitative Finance  Computational Finance;
 Physics  Data Analysis;
 Statistics and Probability;
 Physics  Physics and Society
 EPrint:
 7 pages, 4 figures