Abstract
A new method of constructing financial networks is derived by measuring transfer entropies, which is a newly introduced measure of transferring information. The standard method, constructing a minimum spanning tree from the correlation coefficient matrix of financial time series, fails to illustrate the asymmetric property of influencing and the weight of correlation. In our new approach, we construct the directed and weighted networks of financial systems by calculating transfer entropy between different financial time series. By this effort, not only can the topology of financial systems be described more correctly, but also the research of dynamic properties of financial systems becomes probable.
Users
Please
log in to take part in the discussion (add own reviews or comments).