Describing financial crisis propagation through epidemic modelling on multiplex networks
Publication Date
July 31, 2025
Creator
Abstract
In this thesis we employ various methods from network science, together with epidemic modelling and extreme value theory, to build and analyse financial crisis propagation models. We use stock price, geographical location, and economic sector data for a set of 398 companies to construct multiplex networks and propose a novel framework for modelling financial contagion using an SIR (Susceptible–Infected–Recovered) epidemic model. We compare different shock transmission models and explore their effectiveness in predicting the spread of financial shock during the 2008 financial crisis and the 2020 financial crisis. To enhance the accuracy of our models, we introduce a change point detection method to detect significant changes in historical crisis data and integrate them into our models accordingly, improving their adaptability to major market events. Additionally, we develop a model that prioritizes recent observations under the assumption that they provide a more accurate reflection of current market conditions and trends, assigning greater weight to recent data while reducing the influence of older data. Our findings highlight the importance of the multiplex network structure, differentiating between various transmission pathways, and demonstrate the value of incorporating change points and weighted observations for more accurate predictions of affected companies, sectors and continents. In addition, there is no single model that performs best in all scenarios. Hence, different predictions tasks, whether forecasting the number of infected companies or making company-specific predictions, may require distinct approaches to achieve more accurate results.
Item Type
ethesis
Thesis Type
PhD
Associated Schools / Departments
School of Mathematical Sciences (UK)
eprints ID
80220
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
Bozhidarova, Malvina, 20302453, corrected.pdf
Type
Full-text
Description
Examined. Thesis with completed minor corrections
Size
7.71 MB
Format
Adobe PDF
Checksum (MD5)
6c4dcbe17900dd8794795fa21556bd99