The growing interest of research in econometric methods for systemic risk analysis fostered a rapid development of econometric spillover and network models to monitor the systemic risk in financial systems and improve investment management practices. The thesis contributes to the literature on econometric interconnectedness and investment management by developing new techniques for building models capable to reveal insights on the complex relationships in economic and financial systems. From a methodological viewpoint, the thesis mostly contributes to the statistical and econometric literature on interconnectedness measurement and to the financial one on portfolio management. From an empirical viewpoint, financial applications are offered for both traditional financial markets and the cryptocurrency one, whose relative importance in the global financial system is growing over time. The contributions of this thesis to the literature are developed in seven self contained chapters. Chapter 2 proposes a Vector Error Correction model based spillover methodology to monitor return connectedness and lead-lag relationships of Bitcoin - and more generally financial - market exchanges. Chapter 3 extends the previous study by means of an in-depth analysis of intra-day data. Chapter 4 proposes a methodology to construct a basket based stablecoin whose value is stable over time and resilient to shocks in the currency market. Chapter 5 examines the lead-lag relationship between the European countries’ sovereign CDS and bond market by means of the effective transfer entropy methodology. Chapter 6 introduces an artificial neural network framework for Bitcoin option pricing. Chapter 7 proposes an asset allocation methodology capable to take into account for the systemic risk impounded into network metrics when dealing with portfolio management, applied to the cryptocurrency space. Chapter 8 proposes a methodology based on chaos and dynamical systems theory for non-linear time series forecasting and investment strategy development.

Il crescente interesse della ricerca in metodi econometrici per l’analisi del rischio sistemico ha favorito un rapido sviluppo di modelli di network e spillover econometrici per monitorare il rischio sistemico nei sistemi finanziari e migliorare le pratiche di investment management. La tesi contribuisce alla letteratura sulle misure d’interconnessione e sull’investment management tramite lo sviluppo di nuove tecniche per costriuire modelli capaci di studiare le relazioni complesse che sussistono nei sistemi economico-finanziari. Da un punto di vista metodologico, la tesi contribuisce alla letteratura statistico-econometrica sulle misure d’interconnessione e a quella finanziaria sulla gestione del portafoglio. Da un punto di vista empirico, le applicazioni finanziarie sono condotte sia nell’ambito dei mercati finanziari, che in quello delle criptovalute, la cui importanza nel sistema finanziario globale è attualmente in crescita. I contributi di questa tesi rispetto alla letteratura sono sviluppati in sette capitoli. Il capitolo 2 propone un misure di spillover basate su modelli Vector Error Correction per monitorare l’interconnessione nei rendimenti e la relazione lead-lag delle piattaforme di scambio di Bitcoin – e, più in generale, di strumenti finanziari. Il capitolo 3 estende il precedente studio per mezzo di un’analisi di dati intra-day. Il Capitolo 4 propone una metodologia per costruire, da un paniere di valute, uno stablecoin il cui valore sia stabile nel tempo e resiliente da shock nelle valute. Il capitolo 5 esamina la relazione lead-lag tra i CDS sovrani dei paesi europei e il mercato obbligazionario tramite la metodologia dell’effective transfer entropy. Il capitolo 6 introduce un modello di rete neurale artificiale per il pricing di opzioni sottoscritte sul Bitcoin. Il capitolo 7 propone una metodologia di allocazione del portafoglio capace di tenere in considerazione il rischio sistemico derivante dalle metriche di network, applicata al contesto delle criptovalute. Il capitolo 8 propone una metodologia basata sul concetto di caos e sulla teoria dei sistemi dinamici complessi per la predizione di serie temporali non stazionarie e l’investment Management.

(2021). Interconnessione, Reti e Tecnologie Finanziarie: dal Rischio Sistemico all'Investment Management . Retrieved from http://hdl.handle.net/10446/185933

Interconnessione, Reti e Tecnologie Finanziarie: dal Rischio Sistemico all'Investment Management

PAGNOTTONI, Paolo
2021-07-06

Abstract

The growing interest of research in econometric methods for systemic risk analysis fostered a rapid development of econometric spillover and network models to monitor the systemic risk in financial systems and improve investment management practices. The thesis contributes to the literature on econometric interconnectedness and investment management by developing new techniques for building models capable to reveal insights on the complex relationships in economic and financial systems. From a methodological viewpoint, the thesis mostly contributes to the statistical and econometric literature on interconnectedness measurement and to the financial one on portfolio management. From an empirical viewpoint, financial applications are offered for both traditional financial markets and the cryptocurrency one, whose relative importance in the global financial system is growing over time. The contributions of this thesis to the literature are developed in seven self contained chapters. Chapter 2 proposes a Vector Error Correction model based spillover methodology to monitor return connectedness and lead-lag relationships of Bitcoin - and more generally financial - market exchanges. Chapter 3 extends the previous study by means of an in-depth analysis of intra-day data. Chapter 4 proposes a methodology to construct a basket based stablecoin whose value is stable over time and resilient to shocks in the currency market. Chapter 5 examines the lead-lag relationship between the European countries’ sovereign CDS and bond market by means of the effective transfer entropy methodology. Chapter 6 introduces an artificial neural network framework for Bitcoin option pricing. Chapter 7 proposes an asset allocation methodology capable to take into account for the systemic risk impounded into network metrics when dealing with portfolio management, applied to the cryptocurrency space. Chapter 8 proposes a methodology based on chaos and dynamical systems theory for non-linear time series forecasting and investment strategy development.
6-lug-2021
33
2019/2020
APPLIED ECONOMICS AND MANAGEMENT (AEM)
GIUDICI, PAOLO STEFANO
Pagnottoni, Paolo
File allegato/i alla scheda:
File Dimensione del file Formato  
PhD_thesis_Paolo_Pagnottoni_1.pdf

accesso aperto

Descrizione: Tesi di Paolo Pagnottoni
Versione: Tesi di dottorato
Licenza: Licenza default Aisberg
Dimensione del file 5.87 MB
Formato Adobe PDF
5.87 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10446/185933
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact