This Thesis covers three different topics in the wide area of mathematical finance. Part 1 deals with the application of matrix-variate Wishart processes in derivatives pricing. Chapter 1 is devoted to the study of existing Wishart-based Stochastic Volatility models. Firstly, we discuss the models calibration and, exploiting the distributional properties ofWishart process, we propose efficient model approximations that alleviate the associated computational burden. Further we highlight the constraints that need to be satisfied in order to get a well-defined Wishart process and their impact on pricing performances. Secondly, we present simple and efficient simulation schemes for the asset price trajectories, making use of the exact sampling scheme in [3], that allow to price path-dependent derivatives. Chapter 2 presents a new class of pricing models that extend the application of Wishart processes to the so-called Stochastic Local Volatility (or hybrid) pricing paradigm. This is a very recent approach that is meant to combine the advantages of Local and Stochastic Volatility models. Part 2, based on a research project conducted with Professor Roberto Baviera (Politecnico di Milano) and Paolo Pellicioli, is concerned with the computation of CVA (Credit Valuation Adjustment) in the presence of Wrong Way Risk. Hull-White approach to Wrong Way Risk in the computation of CVA is considered the most straightforward generalization of the standard Basel approach. The model is financially intuitive and it can be implemented by a slight modification of existing algorithms for CVA calculation. However, path dependency in the key quantities has non elementary consequences in the calibration of model parameters. We propose a simple and fast approach for computing these quantities via a recursion formula. We show that the proposed methodology leads to a straightforward application of Hull-White model to the computation of CVA for portfolios of derivatives with early termination features, such as American or Bermuda options. Part 3, based on a research project conducted with Professors Emilio Barucci and Daniele Marazzina (Politecnico di Milano), is devoted to the study of the impact of relative performance based salary schemes on the risk taking incentives of asset managers. We analyze the asset management problem when the manager is remunerated through a scheme based on the performance of the fund with respect to a benchmark. We provide empirical evidence on the mutual fund industry showing excessive risk taking in case of a very poor performance and limited risk taking in case of overperformance with respect to the benchmark
La presente Tesi copre 3 diversi argomenti appartenenti al vasto campo della finanza matematica. La Parte 1 riguarda l’applicazione dei processi stocastici matriciali di tipo Wishart per la valutazione di strumenti derivati. Inizialmente, presentiamo il problema della calibrazione dei modelli già esistenti in letteratura e, sfruttando le proprietà distribuzionali del processo di Wishart, proponiamo delle approssimazioni di modello efficienti che permettono di alleggerire il carico computazionale richiesto. Inoltre, ampio risalto è dato alle condizioni che devono essere imposte al fine di ottenere un processo di Wishart ben definito e all’impatto di dette condizioni sull’accuratezza del modello. In secondo luogo, presentiamo degli algoritmi di simulazione per discretizzare le traiettorie dei prezzi. Infine presentiamo una nuova classe di modelli a volatilità stocastica-locale che generalizzano i modelli preesistenti. La Parte 2, basata su un progetto di ricercar svolto con Roberto Baviera (Politecnico di Milano) e Paolo Pellicioli, verte sula valutazione del CVA (Credit Valuation Adjustment) in presenza di Wrong Way Risk. Il modello proposto da Hull e White in tema di Wrong Way Risk è considerate il più intuitivo dal punto di vista finanziario e può essere implementato tramite semplici modifiche agli algoritmi esistenti. È importante segnalare, però, che il problema della determinazione di alcune quantità chiave del modello è di tipo path dependency con ricadute significative sul processo di calibrazione del modello. Proponiamo un approccio semplice ed efficiente per risolvere questo problema tramite una formula iterativa. Mostriamo, inoltre, come la metodologia da noi proposta conduca in maniera del tutto intuitiva alla valutazione del CVA per portafogli di strumenti derivati che presentano clausole di esercizio anticipato, quali le opzioni Americane o Bermuda. La Parte 3, basata su un progetto di ricerca condotto con Emilio Barucci e Daniele Marazzina (Politecnico di Milano), verte sullo studio dell’impatto delle strutture di remunerazione dei basate sulla performance relativa sugli incentivi alla rischiosità dei gestori di un fondo di investimento. Analizziamo dal punto formale e rigoroso il problema di asset management del gestore e forniamo delle evidenze empiriche di mercato a sostegno dell’analisi effettuata.
Three essays in mathematical finance
LA BUA, GAETANO
Abstract
This Thesis covers three different topics in the wide area of mathematical finance. Part 1 deals with the application of matrix-variate Wishart processes in derivatives pricing. Chapter 1 is devoted to the study of existing Wishart-based Stochastic Volatility models. Firstly, we discuss the models calibration and, exploiting the distributional properties ofWishart process, we propose efficient model approximations that alleviate the associated computational burden. Further we highlight the constraints that need to be satisfied in order to get a well-defined Wishart process and their impact on pricing performances. Secondly, we present simple and efficient simulation schemes for the asset price trajectories, making use of the exact sampling scheme in [3], that allow to price path-dependent derivatives. Chapter 2 presents a new class of pricing models that extend the application of Wishart processes to the so-called Stochastic Local Volatility (or hybrid) pricing paradigm. This is a very recent approach that is meant to combine the advantages of Local and Stochastic Volatility models. Part 2, based on a research project conducted with Professor Roberto Baviera (Politecnico di Milano) and Paolo Pellicioli, is concerned with the computation of CVA (Credit Valuation Adjustment) in the presence of Wrong Way Risk. Hull-White approach to Wrong Way Risk in the computation of CVA is considered the most straightforward generalization of the standard Basel approach. The model is financially intuitive and it can be implemented by a slight modification of existing algorithms for CVA calculation. However, path dependency in the key quantities has non elementary consequences in the calibration of model parameters. We propose a simple and fast approach for computing these quantities via a recursion formula. We show that the proposed methodology leads to a straightforward application of Hull-White model to the computation of CVA for portfolios of derivatives with early termination features, such as American or Bermuda options. Part 3, based on a research project conducted with Professors Emilio Barucci and Daniele Marazzina (Politecnico di Milano), is devoted to the study of the impact of relative performance based salary schemes on the risk taking incentives of asset managers. We analyze the asset management problem when the manager is remunerated through a scheme based on the performance of the fund with respect to a benchmark. We provide empirical evidence on the mutual fund industry showing excessive risk taking in case of a very poor performance and limited risk taking in case of overperformance with respect to the benchmarkFile | Dimensione | Formato | |
---|---|---|---|
Gaetano La Bua - PhD Thesis - Definitive.pdf
accessibile in internet per tutti
Descrizione: Thesis - Definitive
Dimensione
1.69 MB
Formato
Adobe PDF
|
1.69 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/131919