A complex systems approach to financial market analysis: Nonlinearity, regime shifts and early warning indicators
MetadataShow full item record
This item's downloads: 7 (view details)
The history of financial markets over the past century points to the stylised fact that markets build up to a peak and then crash. Many of the standard methods for risk estimation and modelling of financial time series rely on the linear stochastic modelling framework. Under this approach, the interaction of market participants are assumed to be independent and, when taken on aggregate, cancel each other out. In order to better capture the build-up of risk in the financial system, the methods applied in this thesis allow for endogenous dynamical behaviour, caused by the complex interaction of market participants as they react and adapt to the trends and patterns they create at an aggregate level. Endogenous dynamics can lead to the build-up of instabilities in a complex system, pushing it closer to a critical threshold. When the critical point is reached, the system may abruptly switch between alternate equilibria. In effect, a regime shift occurs in the system. We investigate whether we can detect certain universal features of complex systems approaching a regime shift to develop early warning indicators of financial crises. Focusing on sovereign bond and stock market time series in the periods leading up to financial crises, the thesis proposes a number of potential indicators of risk building up in the financial system. In particular, we present evidence of nonlinear dependence structures, critical slowing down, and changing network topology in financial markets in the periods preceding financial crises. Taken on aggregate, the results presented in this thesis indicate that moving beyond the linear stochastic framework and applying methods that can capture complex interactions of market participants, and the resultant emergent patterns they create, adds significant value in the development of early warning signals for financial crises.
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. Please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.
The following license files are associated with this item: