The purpose of this work is to solve the problem of pricing American put options under a Lèvy process. In order to do that, the PIDE is reformulated into a fractional partial differential equation. These are two equivalent equations, but in the second expression the integral term is replaced by fractional derivatives. This is a relevant advantage for determining the fundamental solution of the FPDE, which depends only on the charac- teristic exponent of the process; in this way it is possible to find an analytical solution for the pricing of American options. The price is the sum of two terms: the price of the corresponding European option and the early exercise premium. The formula is eva- luated numerically by using two different methods: the former exploits the fast Fourier transform, while the latter the Gauss-Chebychev points. In the same way is derived the optimal-exercise boundary that is part of the solution itself. By simplifying the solution, an approximated formula is provided for American contract with a short maturity; the main benefit given by this approximation is that it does not require solving for the early exercise premium, which is the most complex term. The work is mainly focused on the CGMY models; starting from Guo and Li study in [19], the work is further extended to log stable. The results can be applied to all the Lévy processes for which a FPDE can be determined.
In questo lavoro si presenta la soluzione del problema di pricing per opzioni put americane, nel caso in cui il sottostante sia un processo di Lévy. Per fare ciò viene introdotta una diversa formulazione della PIDE, la fractional partial differential equation. Si tratta di equazioni equivalenti, ma la FPDE non presenta il termine integrale che viene riformulato tramite fractional derivatives. Questo costituisce un vantaggio notevole perché consente di calcolare la soluzione fondamentale della FPDE, che dipende solamente dall’esponente caratteristico del processo, e di ottenere la soluzione esplicita per il pricing di americane. Il prezzo si compone di due termini, il valore delle corrispondenti europee e l’early exercise premium. Si presentano due diversi metodi per ottenere numericamente la soluzione a partire da questi risultati; la prima sfrutta la fast fourier transform e la seconda i nodi di Gauss Chebychev. In questo modo è possibile calcolare l’optimal exercise boundary, anch’esso parte della soluzione del problema. Effettuando alcune semplificazioni sulla soluzione, si fornisce una approssimazione valida per i contratti americani con scadenze brevi; il pregio di questa approssimazione è quello di non richiedere la soluzione dell’early exercise premium che è il termine più complesso. Ci si è focalizzati principalmente sui modelli CGMY, seguendo il lavoro di Guo e Li in [19]; in aggiunta è stato esteso il lavoro ai processi log stable. Questi risultati possono essere estesi a tutti i Lévy per cui si riesca a ricavare la FPDE.
Pricing di opzioni Americane tramite fractional partial differential equation per CGMY e log stable
CELORA, CHIARA
2016/2017
Abstract
The purpose of this work is to solve the problem of pricing American put options under a Lèvy process. In order to do that, the PIDE is reformulated into a fractional partial differential equation. These are two equivalent equations, but in the second expression the integral term is replaced by fractional derivatives. This is a relevant advantage for determining the fundamental solution of the FPDE, which depends only on the charac- teristic exponent of the process; in this way it is possible to find an analytical solution for the pricing of American options. The price is the sum of two terms: the price of the corresponding European option and the early exercise premium. The formula is eva- luated numerically by using two different methods: the former exploits the fast Fourier transform, while the latter the Gauss-Chebychev points. In the same way is derived the optimal-exercise boundary that is part of the solution itself. By simplifying the solution, an approximated formula is provided for American contract with a short maturity; the main benefit given by this approximation is that it does not require solving for the early exercise premium, which is the most complex term. The work is mainly focused on the CGMY models; starting from Guo and Li study in [19], the work is further extended to log stable. The results can be applied to all the Lévy processes for which a FPDE can be determined.File | Dimensione | Formato | |
---|---|---|---|
Tesi_ChiaraCelora.pdf
solo utenti autorizzati dal 10/07/2020
Dimensione
3.61 MB
Formato
Adobe PDF
|
3.61 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/10589/134924