Scaling properties are among the most important quantifiers of complexity in many real systems, including chaotic dynamical systems, biological systems and financial markets. The presence of scaling usually points out a non-trivial self-organizing behavior and temporal correlations. This thesis reports on a nontrivial use of the Detrended Fluctuation Analysis and its generalizations, like the Generalized Hurst Exponent and its weighted version. We discussed about their practical applicability to datasets originating from completely different fields of research, from economics to physiology. The innovative features of this thesis are twofold: on one side it has been possible to verify that the DFA turned out to be a successful technique even when the dataset is not equally temporally spaced; on the other side, it has been possible to obtain - after the combination of the DFA with a complementary tool of investigation - a discriminating procedure and a classification instrument, well suited for diagnostic and clinical purposes. Moreover the same procedures, once extended to GHE analysis, allowed to characterize the multifractal features and cross-correlation properties in commodities markets in order to identify different scales and temporal horizons in trading procedures. This is of crucial importance for the improvement of reliability of risk management and option pricing models.
Scheda prodotto non validato
Scheda prodotto in fase di analisi da parte dello staff di validazione
|Titolo:||Complex Systems Approaches in Investigating Series of Data: Multidisciplinary Applications in Economics and Physiology|
|Appare nelle tipologie:||8.1 Tesi Dottorato|
File in questo prodotto: