First principles and effective theory approaches to dynamics of complex networks
Lists
Only you see this
Abstract
This dissertation concerns modeling two aspects of dynamics of complex networks: (1) \nresponse dynamics and (2) growth and formation. \nA particularly challenging class of networks are ones in which both nodes and links are \nevolving over time – the most prominent example is a financial network. In the first part \nof the dissertation we present a model for the response dynamics in networks near a metastable \npoint. We start with a Landau-Ginzburg approach and show that the most general \nlowest order Lagrangians for dynamical weighted networks can be used to derive conditions \nfor stability under external shocks. Using a closely related model, which is easier to solve \nnumerically, we propose a powerful and intuitive set of equations for response dynamics \nof financial networks. We find the stability conditions of the model and find two phases: \n“calm” phase , in which changes are sub-exponential and where the system moves to a new, \nclose-by equilibrium; “frantic” phase, where changes are exponential, with negative blows \nresulting in crashes and positive ones leading to formation of "bubbles". We empirically \nverify these claims by analyzing data from Eurozone crisis of 2009-2012 and stock markets. \nWe show that the model correctly identifies the time-line of the Eurozone crisis, and in the stock market data it correctly reproduces the auto-correlations and phases observed in the \ndata. \nThe second half of the dissertation addresses the following question: Do networks that \nform due to local interactions (local in real space, or in an abstract parameter space) have \ncharacteristics different from networks formed of random or non-local interactions? Using \ninteracting fields obeying Fokker-Planck equations we show that many network characteristics \nsuch as degree distribution, degree-degree correlation and clustering can either be \nderived analytically or there are analytical bounds on their behaviour. In particular, we \nderive recursive equations for all powers of the ensemble average of the adjacency matrix. \nWe analyze a few real world networks and show that some networks that seem to form from \nlocal interactions indeed have characteristics almost identical to simulations based on our \nmodel, in contrast with many other networks.
BibTeX
@inproceedings{dehmamy2016first,
author = {Nima Dehmamy},
title = {{First principles and effective theory approaches to dynamics of complex networks}},
booktitle = {OpenBU},
year = {2016},
}Loading related content...