The need for effective simulation techniques, when studying the performance of portfolio investments in financial applications, was recognized since it was observed that backtesting typically introduces significant bias. However, while Monte Carlo simulations are commonly used in this application scenario, up to now no general frameworks have been proposed. This paper describes a general modeling and simulation framework that is used to study how allocation schemes perform when different synthetic time series generation models are employed. Moreover, we devised a novel portfolio allocation scheme where assets are nodes of a complex network and communities of correlated assets are detected and measured by means of modularity. Allocation is than obtained by equally distributing weights among different communities. We compare this novel scheme against state-of-the-art approaches in various scenarios, under Gaussian, Geometric Brownian motion and ARFIMA generation models. Results show that the proposed scheme outperforms the others in many scenarios.

Network Modularity based Clustering for Portfolio Allocation: a Monte-Carlo Simulation Study

Ferretti, Stefano
Conceptualization
;
Montagna, Sara
Membro del Collaboration Group
2022

Abstract

The need for effective simulation techniques, when studying the performance of portfolio investments in financial applications, was recognized since it was observed that backtesting typically introduces significant bias. However, while Monte Carlo simulations are commonly used in this application scenario, up to now no general frameworks have been proposed. This paper describes a general modeling and simulation framework that is used to study how allocation schemes perform when different synthetic time series generation models are employed. Moreover, we devised a novel portfolio allocation scheme where assets are nodes of a complex network and communities of correlated assets are detected and measured by means of modularity. Allocation is than obtained by equally distributing weights among different communities. We compare this novel scheme against state-of-the-art approaches in various scenarios, under Gaussian, Geometric Brownian motion and ARFIMA generation models. Results show that the proposed scheme outperforms the others in many scenarios.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11576/2705378
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact