Despite these advantages,traditional feedback identification
theories often suffer from the opinion that they usually address
two-variate time-series data and are inappropriate for large-scale
networks because of their practical and theoretical
limitations.Data acquisition is difficult or connective
entanglements are fearing,which might hinder their applications to
very large datasets, as occur more and more frequently nowadays.
Now this is not the case,as many new experimental techniques, for
example,real-time PCR,
immunofluorescence,microarray,multi-electrode array and EEG;can now
provide such time-series data in a cost efficient manner. Also a
multi-variate time-series analysis theory has undergone a great
development.The new theoretical contribution much helps to find the
feedback loops in large-scale networks.