This thesis aims to show that in some applications the appropriate selection of a small number of available items can be beneficial with respect to the use of all available items. In particular, we focus on portfolio selection and on operational risk management and we use operations research techniques to identify the few important elements that are needed in both cases. In the first part of this work - based on an article published in Economics Bulletin [Cesarone et al (2016)], we show that, for several portfolio selection models, the best portfolio which uses only a limited number of assets has in-sample performance very close to that of an optimized portfolio which could include all assets, but generally obtains better out-of-sample performance. This is true for various performance measures, and it is often possible to identify a "golden range" of sizes where the best performances are obtained. These general empirical findings are consistent with theoretical results obtained by Kondor and Nagy (2007) under very restrictive assumptions. We also note that small portfolios are preferable for several practical reasons including monitoring, availability for small investors, and transaction costs. In the second part of the thesis, we develop an operational risk management framework for the assessment of the exposure of a company (with particular reference to a financial institution) to potential risk events arising from the launch of a new product. This framework is based on the Analytic Hierarchy Process and on the 80/20 rule which allows one to rank and to identify the most relevant risk events, respectively. By means of appropriate integer programming models we then address the problem of identifying the mitigation actions that secure the internal processes of a company with minimum cost. This corresponds to the primary goal of an operational risk manager: reducing the exposure to potential risk events. An alternative approach, when the budget is fixed, consists in selecting the subset of mitigation actions that provide the greatest reduction in operational risk exposure for that budget. A parametric analysis with respect to the budget level provides additional information for the management to take decisions about possible budget adjustments.

Fewer is better: the cases of portfolio selection and of operational risk management / Moretti, Jacopo. - (2018 Feb 01).

Fewer is better: the cases of portfolio selection and of operational risk management

MORETTI, JACOPO
01/02/2018

Abstract

This thesis aims to show that in some applications the appropriate selection of a small number of available items can be beneficial with respect to the use of all available items. In particular, we focus on portfolio selection and on operational risk management and we use operations research techniques to identify the few important elements that are needed in both cases. In the first part of this work - based on an article published in Economics Bulletin [Cesarone et al (2016)], we show that, for several portfolio selection models, the best portfolio which uses only a limited number of assets has in-sample performance very close to that of an optimized portfolio which could include all assets, but generally obtains better out-of-sample performance. This is true for various performance measures, and it is often possible to identify a "golden range" of sizes where the best performances are obtained. These general empirical findings are consistent with theoretical results obtained by Kondor and Nagy (2007) under very restrictive assumptions. We also note that small portfolios are preferable for several practical reasons including monitoring, availability for small investors, and transaction costs. In the second part of the thesis, we develop an operational risk management framework for the assessment of the exposure of a company (with particular reference to a financial institution) to potential risk events arising from the launch of a new product. This framework is based on the Analytic Hierarchy Process and on the 80/20 rule which allows one to rank and to identify the most relevant risk events, respectively. By means of appropriate integer programming models we then address the problem of identifying the mitigation actions that secure the internal processes of a company with minimum cost. This corresponds to the primary goal of an operational risk manager: reducing the exposure to potential risk events. An alternative approach, when the budget is fixed, consists in selecting the subset of mitigation actions that provide the greatest reduction in operational risk exposure for that budget. A parametric analysis with respect to the budget level provides additional information for the management to take decisions about possible budget adjustments.
1-feb-2018
File allegati a questo prodotto
File Dimensione Formato  
Tesi dottorato Moretti

accesso aperto

Tipologia: Tesi di dottorato
Licenza: Creative commons
Dimensione 3.89 MB
Formato Adobe PDF
3.89 MB Adobe PDF

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/11573/1125508
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact