The authors perform analysis of theoretical and empirical research on financial networks and disclose the peculiarities of network analysis application for evaluation of systemic financial integration. The analysed methods are tested on international portfolio investments network.

The connectivity among different economic agents (persons, organizations, countries) is constantly growing and their activity become more and more interdependent. Social, informational, ownership, trade, investment and other ties become closer and this determines such processes as trade and financial integration, synchronization of business cycles and othe processes of interdependence. Connectivity processes are conditioned by such factors as: market liberalization, increasing global competition, technological and informational innovations, increasing speed of information distribution, outsourcing processes and etc. The relevance, scope, threats and opportunities of economic agents connectivity were revealed by the economic crisis of 2008. It also exposed that traditional means of analysis and forecasting are no longer capable to adequately evaluate the complexity of economic agents and their performance. Economists try to solve this problem by applying network analysis to economic research. Network analysis is not an invetion of economic scientists, but adopted from other disciplines. In order for this addoption process to be successful, there should be disclosed the peculiarities of network analysis application to economic research. After a thorough analysis of the existing theoretical and empirical research in this field, the authors of this article identified four groups of peculiarities that determine the possibilities to apply network analysis to economic research: historic background, network elements, dimensions of analysis and fields of application. The frontiers of research examining economic networks have been advancing along two strands: one emanating from economics and sociology, the other from research on complex systems in physics and computer science.[...].