Current Students

COMPG009 - Networks and Systemic Risk

This database contains 2016-17 versions of the syllabuses. For current versions please see here.

CodeCOMPG009 
YearMSc
PrerequisitesFamiliarity with basic probability and calculus; the coursework requires basic programming skills
Term2
Taught ByFabio Caccioli (100%)
Aims

The first part of the course presents a general introduction to complex networks and dynamical processes. The second part is focused on specific applications to the study of contagion in financial networks. Overall, the course represents an introduction to the topic of systemic risk and stress propagation in networked systems.

Learning Outcomes

The student will be able to 1) compute network metrics and provide a statistical description of networks, 2) analyze dynamical processes on networks, 3) implement simple algorithms for the analysis of financial contagion. 

Content

Introduction to complex networks 

  • Basic concepts of networks (graphs, subgraphs, adjacency matrix, undirected, directed and weighted networks), common metrics (degree, betweenness, centrality, clustering, degree distribution, excess degree distribution, mixing patterns, real world examples). 
  • Network models (random networks, configuration model, small world, preferential attachment). 

  

Collective behavior 

  • Emergence of a giant cluster. Robustness to random and targeted attacks.
  • Spreading processes, master equation, approximate solution of the master equation, mean field approximation, computer simulations.
  • Cascade processes on networks.

  

Application to interbank networks and systemic risk

  • Interbank lending networks, payment networks, and their properties.
  • Furfine default algorithm and cascades of defaults.
  • Clearing vector of payments and the Eisenberg-Noe model.

Method of Instruction

30 hours of lectures plus homework and assignments

Assessment

The course has the following assessment components:

  • Coursework (50%, individual project on financial networks and written essay)
  • Written examination (2.5 hours, 50%)

To pass this course, students must:

 

  • Obtain an overall pass mark of 50% for all sections combined.

Resources

Recommended texts
Barrat, Bathelemy and Vespignani “Dynamical processes on complex  networks”, Cambridge University Press
Newman “Networks: an introduction”, Oxford University Press 

Further reading
Barabasi and Albert, “Statistical mechanics of complex networks ”Review of Modern Physics 74, 47
Lorenz, Battiston and Schweitzer“Systemic Risk in a Unifying Framework for Cascading Processes on Networks” European Journal of Physics B 71, no 4 441-460